Selective and brain-permeable polo-like kinase-2 (Plk-2) inhibitors that reduce α-synuclein phosphorylation in rat brain

CHEMMEDCHEM

FULL PAPERS

Gets into your head: By using a struc-

ture-guided drug discovery approach,

highly selective brain-penetrant Plk-2 in-

hibitors were designed with the use of

an interesting aromatic edge–face inter-

action as a potency–selectivity determi-

nant. An analogue from this work low-

ered phosphorylated a-synuclein levels

in vivo on oral dosing, demonstrating

successful target engagement in the rat

brain and paving the way for proof-of-

concept studies in rodent models of

Parkinson’s disease.

D. L. Aubele, R. K. Hom, M. Adler,

R. A. Galemmo, Jr., S. Bowers,

A. P. Truong, H. Pan, P. Beroza, R. J. Neitz,

N. Yao, M. Lin, G. Tonn, H. Zhang,

M. P. Bova, Z. Ren, D. Tam, L. Ruslim,

J. Baker, L. Diep, K. Fitzgerald, J. Hoffman,

R. Motter, D. Fauss, P. Tanaka, M. Dappen,

J. Jagodzinski, W. Chan, A. W. Konradi,

L. Latimer, Y. L. Zhu, H. L. Sham,

J. P. Anderson, M. Bergeron, D. R. Artis*

&& – &&

Selective and Brain-Permeable Polo-

like Kinase-2 (Plk-2) Inhibitors That

Reduce a-Synuclein Phosphorylation

in Rat Brain

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 20

&

1

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

DOI: 10.1002/cmdc.201300166

Selective and Brain-Permeable Polo-like Kinase-2 (Plk-2)

Inhibitors That Reduce a-Synuclein Phosphorylation in Rat

Brain

Danielle L. Aubele,^[a]Roy K. Hom,^[a]Marc Adler,^[a]Robert A. Galemmo, Jr.,^{[a, d]}

Simeon Bowers,^[a]Anh P. Truong,^[a]Hu Pan,^[a]Paul Beroza,^[a]R. Jeffrey Neitz,^{[a, e]}Nanhua Yao,^[a]

May Lin,^[a]George Tonn,^[b]Heather Zhang,^[b]Michael P. Bova,^[a]Zhao Ren,^[a]Danny Tam,^[a]

Lany Ruslim,^[a]Jeanne Baker,^[a]Linnea Diep,^[a]Kent Fitzgerald,^[b]Jennifer Hoffman,^[b]

Ruth Motter,^[b]Donald Fauss,^[a]Pearl Tanaka,^[b]Michael Dappen,^[a]Jacek Jagodzinski,^[a]

Wayman Chan,^[a]Andrei W. Konradi,^[a]Lee Latimer,^[a]Yong L. Zhu,^[a]Hing L. Sham,^[a]

John P. Anderson,^[c]Marcelle Bergeron,^[b]and Dean R. Artis*^[a]

Polo-like kinase-2 (Plk-2) has been implicated as the dominant

kinase involved in the phosphorylation of a-synuclein in Lewy

bodies, which are one of the hallmarks of Parkinson’s disease

neuropathology. Potent, selective, brain-penetrant inhibitors of

Plk-2 were obtained from a structure-guided drug discovery

approach driven by the first reported Plk-2–inhibitor com-

plexes. The best of these compounds showed excellent iso-

form and kinome-wide selectivity, with physicochemical prop-

erties sufficient to interrogate the role of Plk-2 inhibition

in vivo. One such compound significantly decreased phosphor-

ylation of a-synuclein in rat brain upon oral administration and

represents a useful probe for future studies of this therapeutic

avenue toward the potential treatment of Parkinson’s disease.

Introduction

Lewy Bodies are intracytoplasmic inclusions found in dopami-

nergic neurons and are recognized as the prime histological

marker of Parkinson’s disease. They are prognostic of Parkin-

son’s disease dementia, dementia with Lewy bodies, multiple

system atrophy, and Lewy body dysphagia.^[1]a-Synuclein is the

major component of Lewy bodies,^[1]where it is found to be

highly phosphorylated at serine 129.^[2]Our research group has

demonstrated,^[3]and others have confirmed,^[4]that polo-like

kinase-2 (Plk-2) is the major enzyme responsible for phosphor-

ylation of a-synuclein at serine 129, thereby suggesting that

Plk-2 may play a key role in the progression of Parkinson’s-like

pathologies. To examine this hypothesis, we aimed at creating

highly selective inhibitors of Plk-2 for use as chemical probes

in various cellular and in vivo models of a-synucleinopathy.^[5]

A

robust co-crystallization system was obtained, leading to the

first reported small-molecule complexes with the Plk-2 kinase

domain. This enabled a computationally assisted, structure-

guided approach that yielded a series of compounds with ex-

cellent kinome selectivity, including selectivity for Plk-2 over

Plk-1 and Plk-3. Many of these compounds displayed reasona-

ble cellular potency, and optimization yielded 21, which had

suitable physicochemical and pharmacokinetic properties to

demonstrate in vivo decreases in phospho-S129 a-synuclein

levels in rat brain upon oral dosing.

[a] Dr. D. L. Aubele, Dr. R. K. Hom, Dr. M. Adler, Dr. R. A. Galemmo, Jr.,

Dr. S. Bowers, Dr. A. P. Truong, Dr. H. Pan, Dr. P. Beroza, Dr. R. J. Neitz,

Dr. N. Yao, M. Lin, Dr. M. P. Bova, Dr. Z. Ren, D. Tam, L. Ruslim, Dr. J. Baker,

L. Diep, D. Fauss, Dr. M. Dappen, Dr. J. Jagodzinski, W. Chan,

Dr. A. W. Konradi, Dr. L. Latimer, Dr. Y. L. Zhu, Dr. H. L. Sham, Dr. D. R. Artis

Molecular Discovery, Elan Pharmaceuticals

180 Oyster Point Boulevard, South San Francisco, CA 94080 (USA)

E-mail: drartis@comcast.net

[b] Dr. G. Tonn, H. Zhang, Dr. K. Fitzgerald, J. Hoffman, R. Motter, P. Tanaka,

Dr. M. Bergeron

Pharmacological Sciences, Elan Pharmaceuticals

180 Oyster Point Boulevard, South San Francisco, CA 94080 (USA)

The polo-like kinase family is characterized by the presence

of the polo-box domain, a conserved motif with roles in pro-

tein–protein binding and modulation of the kinase domain ac-

tivity.^[6]This family has drawn a great deal of interest for its

varied and critical roles in control of the cell cycle.^[6,7]Plk-1 is

the most studied as a potential target for cancer therapeu-

tics,^[8]and X-ray crystal structures have been published for the

Plk-1 kinase domain.^[9]The functions remain less clear for the

other members of the family: Plk-2 (or serum-inducible kinase,

Snk), Plk-3, Plk-4, and particularly the recently identified Plk-5,

which lacks a functional kinase domain.^[10]

[c] Dr. J. P. Anderson

Department of Biology, Elan Pharmaceuticals

180 Oyster Point Boulevard, South San Francisco, CA 94080 (USA)

[d] Dr. R. A. Galemmo, Jr.

Current address: Southern Research Institute

2000 Ninth Avenue South, Birmingham, AL 35205 (USA)

[e] Dr. R. J. Neitz

Current address: Small-Molecule Discovery Center

University of California San Francisco, San Francisco, CA 94158 (USA)

Supporting information for this article is available on the WWW under

http://dx.doi.org/10.1002/cmdc.201300166.

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 20

&

2

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

Neither Plk-3 nor Plk-2 function is essential, as knockout

mice are viable.^[11]In addition to a-synuclein phosphorylation

and a putative role in neurodegenerative pathology, Plk-2 is in-

volved in modulation of cell division in the presence of micro-

tubule disruption^[12]or mitochondrial impairment.^[13]In con-

junction with Cdk-5, Plk-2 plays a critical role in neuronal ho-

meostatic plasticity.^[14]It is involved in control of excitation-de-

pendent synaptic plasticity resulting in destabilization of post-

synaptic structures. This follows the binding or

phosphorylation of targets including the spine-associated rap

(SPAR) guanosine triphosphatase activating protein^[15]and N-

ethylmaleimide-sensitive fusion protein (NSF).^[16]The roles of

Plk-3 include regulation of cell-cycle checkpoints,^[17]modula-

tion of the response to hypoxia,^[11b]and synaptic destabiliza-

tion.^[18]The contribution of Plk-3 to a-synuclein phosphoryla-

tion is less clear than for Plk-2; although the in vitro activity of

isolated Plk-3 is similar to that of Plk-2, it has not been as ex-

tensively validated in vivo.^[3,4]

The availability of selective, high-affinity chemical probes for

the Plks would be of obvious utility to elucidate their role in

cellular functions. For our intended target validation studies,

a chemical probe inhibitor with high kinome-wide specificity

and isoform selectivity for Plk-2 over the other Plks would be

required. To be a useful tool for evaluating the role of Plk-2 in

cultured cells and in vivo, it would also be necessary to avoid

potentially toxic effects of Plk-1 inhibition in dividing cells, and

selectivity for Plk-2 over Plk-3 would eliminate the latter as

a potentially confounding mechanistic element. For the evalua-

tion of in vivo activity, a metabolically stable compound that

can achieve good CNS exposures in rodents would be re-

quired. Our efforts resulted in two closely related series of

compounds that have enabled in vivo studies yielding encour-

aging data, and a portion of the compounds derived from this

work, resulting from a focused sub-series SAR expansion, have

been described separately.^[19]

Figure 1. a) Biochemical potency of 1 against Plk-1–3. b) Crystal structure of

1 bound to Plk-2 (PDB code 4I5M, 1.8 ꢁ resolution).

Based on this information, it appeared likely that with suffi-

cient structural information, selective, high-affinity analogues

targeting Plk-2 could be obtained from the pteridinone scaf-

fold. We were particularly interested in exploiting the Leu-to-

Tyr difference at position 161 that differentiates Plk-2 from Plk-

1 and -3, and reasoned that exploiting this change could pro-

vide the means to obtain isoform-selective analogues. To this

end, a crystal structure of 1 with the kinase domain of Plk-2

was obtained (Figure 1b). This crystal structure represented

the first published structure for Plk-2 (PDB code 4I5M). Many

of the binding interactions corresponded to contacts reported

previously^[21]for 1 with Plk-1 (PDB code 2RKU). Critically, the

central amide in the hinge, Cys162, donates a hydrogen bond

to the pteridinone N1 atom (2.9 ꢁ). This mimics the interaction

of the N1 purine of adenosine with Plk-1 (PDB code 2OU7).

There were several other interactions that mimic the binding

of 1 to Plk-1 and likely contribute to the observed selectivity

against a broad range of kinases: 1) the 5-methyl group is

packed against the gatekeeper Leu159, 3.9 ꢁ away; 2) the 6-

carbonyl tied into the water network in the specificity pocket;

3) the 7-ethyl substituent is in close contact with Cys96 (Sg,

3.9 ꢁ) in the P-loop; 4) the 8-cyclopentane ring induces a con-

formational change in the P-loop that results in a close contact

with the backbone of Lys90 (4.2 ꢁ); and 5) the pteridinone ring

is packed against the side chain of Phe212 (3.5 ꢁ).

Results and Discussion

The known Plk inhibitor BI 2536 (1),^[8,20]a clinical oncology

pteridinone with a moderate level of selectivity over other

kinase families and IC₅₀values in the low nanomolar range for

Plk-1–3 (Figure 1a), provided an excellent starting point for the

development of Plk-2-specific inhibitors.^[21]The co-crystal struc-

ture of Plk-1 with 1 in the ATP binding site has been reported,

along with analysis of the important binding contacts and a hy-

pothesis for the observed kinase selectivity and potency. The

location of the ortho-methoxy substituent of the aniline side

chain of 1 in a pocket created by the gatekeeper+2 residue

Leu132 served as a steric constraint that attenuated activity for

kinases with a larger Phe or Tyr at this position. The reported

selectivity data at 1 mm concentration of 1 against a panel of

21 kinases supported this model. The only exception was Plk-2,

a Plk isoform with Tyr at the gatekeeper+2 position. The au-

thors argued that 1 was active against Plk-1–3 because the re-

lated active sites provided many similar productive interactions

with 1 across all three isoforms.^[21]

The most notable difference between Plk-1 and Plk-2

around the compound involves the aforementioned Leu–Tyr

substitution in the middle of the hinge (residues Leu132 and

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 20

&

3

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

Tyr161, respectively). Due to a slight shift in the orientation of

residues adjacent to Tyr161, the hinge backbone is further

from the inhibitor than in Plk-1, with a maximum displacement

of 0.7 ꢁ at the Tyr161 Ca carbon position. This results in small

adjustments of the two hydrogen bonds to the hinge at

Cys162. In addition, while the ortho-methoxy group in 1 is

fairly well packed against Leu132 in Plk-1, it exhibits a steric

clash with the side chain of Tyr161 in Plk-2 (3.3 ꢁ). This likely

contributes to the ~10-fold loss in potency against Plk-2 rela-

tive to that observed against Plk-1.

Based on this analysis, we formed the hypothesis that the in-

teraction of the pteridinone 1a with Plk-2 would be improved

by forming an edge-to-face interaction with the aryl side chain

of Tyr161. Modeling studies (see below) suggested that either

a five- or six-membered ring attached to the 2-position

through a biaryl linkage could be accommodated. To test this

hypothesis we synthesized the series of heterocyclic analogues

listed in Table 1. The 2-pyrrole ring analogue 4 has improved

potency and selectivity relative to 1a, and was the first to be

examined structurally. A crystal structure of 4 in Plk-2 (PDB

code 4I5P) revealed a clear edge-to-face^[22]interaction between

the 2-pyrrole group and Tyr161, indicated by: 1) close contact

(3.5 ꢁ) between the C3/C4 edge of the pyrrole moiety and the

4-hydroxyphenyl side chain of Tyr161; 2) the fact that the

alignment of the edge of the pyrrole ring with the center of

the hydroxyphenyl group of Tyr161 resides in an “offset T” rela-

tionship, consistent with the energy minima for an edge-to-

face interaction predicted by benzene dimer model calcula-

tions.^[23]However, the shallow approach angle (588) of the pyr-

role ring in 4 to the Tyr161 hydroxyphenyl group was a sub-

stantial deviation from the optimal 908 angle predicted by this

model. The five-membered pyrrole ring of 4 was nearly copla-

nar with the pteridinone, perhaps due to the intramolecular

hydrogen bond between the pyrrole NH and N1 of the pteridi-

We reasoned that the portion of the inhibitor near Tyr161 is

likely not optimally engineered for Plk-2 potency. To examine

this further, we used compound 1a, a fragment of 1 consisting

of the pteridinone core without the aniline side chain. Al-

though this results in a loss of more than half the atoms in the

inhibitor, the potency of this compound is decreased by only

130-fold. In addition, compound 1a shows equipotent activity

against both Plk-2 and Plk-1, representing a 10-fold improve-

ment in relative selectivity and supporting the idea that the

ortho-methoxy substituent represents a negative binding de-

terminant for Plk-2. The crystal structure of Plk-2 with com-

pound 1a revealed that the pteridinone core maintains almost

all of the binding interactions of the larger compound (Fig-

ure 2a, PDB code 4I6B).

Figure 2. Crystal structures of four inhibitors bound to Plk-2. a) Compound 1a: PDB code 4I6B, 1.8 ꢁ resolution; the protein is shown with cyan carbon

atoms, compound 1a in green, and the superimposed structure of 1 in pink. b) Compound 4: PDB code 4I5P, 1.7 ꢁ resolution; the edge-to-face angle for the

Tyr161/pyrrole interaction is highlighted with an orange line. c) Compound 9: PDB code 4I6F, 2.9 ꢁ resolution, showing the corresponding edge-to-face angle.

d) Compound 11: PDB code 4I6H, 1.9 ꢁ resolution.

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 20

&

4

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

Table 1. Enzyme inhibition data and Plk isoform selectivity ratios for compounds 1a–7.

Compd

R¹

IC₅₀[mm]^[a]

Plk-1

Selectivity

Plk-2/Plk-1

MetStab

(m,r,h)^[b]

clogP^[c]

Plk-2

1.17

1.24

Plk-3

1a

2

Hꢀ

1.45

13.2

–

1.1

0, 1, 19

0, 0, 0

1.9

2.6

10.6

3

4

0.061

0.377

0.306

1.21

18.618

–

5.1

3.2

66, 5, 85

0, 0, 12

2.5

2.8

5

6

7

0.10

0.253

0.167

0.551

–

2.53

5.4

12, 11, 52

0, 0, 22

2.3

2.2

2.8

0.031

0.071

1.464

10.772

7.7

4, 5, 20

[a] Values are means of at least two experiments at 100 mm ATP concentration; details can be found in the Supporting Information. [b] See ref. [23]. [c] De-

tails regarding the method of calculation can be found in the Supporting Information.

none. The rigidity of the two aromatic ring systems affect both

the selectivity and potency of 4. Consequently, our focus was

placed on identifying a replacement group that could exhibit

more optimal interactions with Tyr161.

inhibited only three kinases by >40%, and no kinase showed

>50% inhibition, suggesting that this design approach was

not inherently likely to result in a dramatic loss of overall

kinome selectivity.

The biochemical data for the initial series of seven heterocy-

clic analogues are listed in Table 1. Considering that 1a

showed Plk-2 IC₅₀=1.17 mm, any derivative in this set with

a substantial improvement in Plk-2 inhibition relative to 1a

was likely due to the interaction added by the appended ring.

The 5-membered alkyl N-pyrrolidine ring analogue 2 represent-

ed a variation that approached the activity of 1a, but did not

contribute any favorable binding energy. A series of 5-mem-

bered aryl heterocyclic analogues 3–7 proved to be potent

modifications, with Plk-2 IC₅₀values of 0.377 mm or better. The

most potent of these was the 2-(thiazo-5-yl) analogue 6, with

an eightfold increase in potency relative to 1a.

The selectivity ratios for Plk-2 versus Plk-1 varied from 3- to

11-fold amongst the inhibitors shown in Table 1, the best rep-

resenting a 100-fold swing in relative potency between Plk-2

and Plk-1 from the starting compound. Plk-3 activity was

almost always significantly worse than that for Plk-1, and thus

eliminated the issue of potentially confounding biological ac-

tivity from the landscape of likely optimization issues. Measure-

ments were performed at a single concentration (10 mm) using

a panel of 35 kinases chosen to represent a broad sampling of

the kinome (Supporting Information table 1). The initial lead, 1,

was included for comparison. Both 1 and 5 demonstrated poor

selectivity by inhibiting at least 10 kinases in the panel by 40–

100%. The rates of inhibition exceeded 80% for at least three

kinases. Kinome selectivity was significantly improved by re-

placing the 4-pyrazole in 5 with a 4-pyridine in 7. Compound 7

The microsomal stability was measured for all of these inhib-

itors, which generally exhibited poor to moderate oxidative

stability across species (mouse, rat, and human).^[24]Because the

inhibitors 1a–7 all showed low lipophilicity (clogP<3), it was

surmised that this liability might be due to site-specific oxida-

tion. In the case of the heteroaryl-substituted analogues, this

likely meant oxidation at the carbon atom adjacent to a heter-

oatom within the pendant aromatic ring.

However, among these compounds, imidazole 3 demon-

strated a reasonable selectivity profile, with no inhibition great-

er than 30% for any kinase outside the Plk family, good stabili-

ty in mouse and human microsomes (but not in rat micro-

somes), and moderate activity in the Plk-2 cell assay (IC₅₀=

4.5 mm). Based on these results, as well as its good solubility

and permeability, compound 3 was selected for further ana-

logue-generating efforts.

Throughout this process of design, synthesis, and biochemi-

cal and structural characterization, molecular dynamics (MD)

simulations were performed to provide a computational pre-

diction of inhibitor binding and complex structure. Once stable

simulation systems had been obtained, molecular mechanics

Poisson–Boltzmann surface area (MM-PBSA) methods^[25]were

used to estimate binding energies for new analogues. General-

ly, reasonable agreement was obtained between computed

and experimental structures (Figure 3) and binding affinities

(the latter will be the subject of a more detailed report). Com-

pound designs with the most favorable binding energies were

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 20

&

5

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

teraction with the side chain of Arg165 (Cg, 4.0 ꢁ). Surprisingly,

and somewhat uniquely in the series, 9 did not demonstrate

any selectivity between Plk-1 and Plk-2. Plk-3 activity was still

more than 10-fold worse than that for Plk-1. Although it is pos-

sible this is partly due to bottom-end compression of the assay

due to the Plk-2 concentration used (the final protein concen-

tration is 3 nm, Supporting Information), we did not have the

means to pursue this hypothesis further. Because small

changes to the molecule resulted in compounds with signifi-

cant isoform selectivity, our analysis suggested that a practical

approach involving fine tuning of the cooperative interaction

between the Tyr161/2’-(phenyl)imidazole and Arg165/2’-(phe-

nyl)imidazole/8-cyclopentane ensemble could hold the key to

improving selectivity among the Plks.

In compound 11, a 4-thiazole ring replaces the phenyl

group of 9. The compound maintained the same potency

against Plk-2, but had an eightfold improvement in isoform se-

lectivity (Table 2). The crystal structure of Plk-2–11 (Figure 2d)

shows a close contact between the sulfur atom in the thiazole

and the side chain His169 (3.7 ꢁ). Plk-1 has Glu140 in the corre-

sponding position. Presumably the thiazole ring of 11 under-

goes more favorable interactions with the basic His169 in Plk-2

than Glu140 in Plk-1. Variation of the 8-cyclopentane ring (8–

17, Table 2) yielded compounds with good potency and im-

proved selectivity for Plk-2 over Plk-1. Decreasing the size of

the 8-substitutent to isopropyl (12) and to cyclobutyl (13) re-

tained biochemical potency and increased the isoform selectiv-

ity between the Plks compared with compounds 3 and 9. Re-

placing the cyclopentane ring of 9 with a tetrahydrofuran-3-yl

ring gave a pair of diastereomers, 14 and 15.^[26]Diastereomer

15 was selected for further diversification due to its excellent

Plk-2 biochemical activity and Plk isoform selectivity. The 3-

fluoro analogue 16 displayed in vitro potency and selectivity

similar to 15. However, 4-fluoro 17 represented a 200-fold im-

provement in isoform selectivity over 9. Fluorination at the

para-position of the 2-(phenyl)imidazole confers microsome

stability on 17 relative to the meta-fluorophenyl isomer 16. To

estimate the selectivity of the tetrahydrofuran-3-yl analogues,

the meta-fluorophenyl analogue 16 was characterized in the

35-kinase panel at 10 mm concentration (Supporting Informa-

tion table 1). Compound 16 did not inhibit any of the non-Plk

kinases by >40%.

Figure 3. Comparisons of X-ray structures and computational models. X-ray

structures are shown with green carbon atoms, and models are colored in

cyan (including solvent atoms). The coordinates for the model structure

were obtained by hierarchical clustering of the MD frames using the ligand

heavy atom coordinates. The MD frame shown is the best representative

(i.e., cluster medoid) of the largest cluster: a) compound 4; b) compound 9.

prioritized for synthesis, unless otherwise driven by a critical

hypothesis or rationale.

Analysis of the structure and subsequent modeling studies

highlighted the presence and potential utility of a small cleft

below the imidazole 2-position of compound 3, tucked inside

the Ca–Cb of Arg165 (Figure 2), sufficient to accommodate

a five- or six-membered ring. Appropriate substitution at the 2-

position of the imidazole enhanced the activity in this series

(Table 2). MM-PBSA calculations suggested that appropriate

substitution at this site would enhance Plk-2 potency. With

a simple methyl substitution in 8, there was a loss in potency

with respect to 3, but increased selectivity and microsomal sta-

bility. However, the addition of a 2’-phenyl ring in 9 provided

a 12-fold boost in potency relative to 3. Examination of 9 in

the kinase panel suggested broad kinome selectivity: with the

exception of the Plks, 9 did not display any significant kinase

inhibition >40% at a concentration of 10 mm (Supporting In-

formation table 1), suggesting a selectivity window of roughly

1000-fold or more. The crystal structure (Figure 2c, PDB code

4I6F) revealed that compound 9 had a more optimal edge-to-

face interaction with Tyr161 (3.5 ꢁ, ] 798). The change in

angle reflected the internal stacking of the additional 2’-phenyl

substituent against the 8-cyclopentane group. This stacking in-

creased the dihedral angle between the pteridinone and the

directly connected imidazole ring (318 in 9 versus 128 for 4).

The 2’-phenyl substituent ring also exhibited a hydrophobic in-

Cellular activity against phosphorylation of a-synuclein for

many of these analogues was also determined in a Plk-2-ex-

pressing HEK293 cell line (Table 2). Given the concentration of

ATP in the enzyme assay (100 mm), a shift of ~10-fold was ex-

pected. However, in some cases an attenuation in cellular po-

tency by two orders of magnitude relative to the biochemical

activity was observed, suggesting some additional factor is ex-

erting an influence.

Microsomal stability of the compounds 8–17 was roughly in-

versely proportional to their lipophilicity, as indicated by clogP

values. While compound 9 exhibited very good potency

against Plk-2 relative to 8, its nearly two-log-unit jump in

clogP was reflected by its extremely poor microsomal stability

across species. Altering the placement of the phenyl group at-

tenuated potency (analogue 10). Switching from a pendent cy-

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 20

&

6

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

Table 2. Enzyme inhibition data and Plk isoform selectivity ratios for compounds 8–17.

Compd

R¹

R²

IC₅₀[mm]^[a]

Plk-1

Selectivity

Plk-2/Plk-1

EC₅₀[mm]

HEK293

MetStab

(m,r,h)^[b]

clogP^[c]

Plk-2

0.56

Plk-3

–

8

4.24

7.6

1.6

–

39, 21, 66

0, 0, 0

3.4

9

0.005

0.008

0.095

0.332

5.2

10

11

0.466

0.005

0.195

0.064

12.981

0.571

0.42

–

51, 5, 26

1, 0, 8

4.7

4.5

13

0.837

12

13

14

15

0.025

0.007

0.021

0.008

0.875

0.194

0.481

0.386

2.019

35

28

23

48

2.99

0.667

–

12, 0, 14

4.6

4.7

3.7

–

1, 0, 3

8.602

1.465

27, 0, 10

51, 13,17

0.830

16

17

0.007

0.018

0.390

3.75

4.593

56

1.20

2.10

35, 1, 5

3.7

4.0

10.401

208

63, 67, 58

[a] Values are means of at least two experiments at 100 mm ATP concentration; details can be found in the Supporting Information. [b] See ref. [23]. [c] De-

tails regarding the method of calculation can be found in the Supporting Information.

clopentyl ring at the 8-position of the pteridinone core to

either a pendent isopropyl group, 12, or cyclobutyl group, 13,

modestly improved the metabolic stability. With the exception

of 11, the addition of heteroatoms (14–17) and blocking po-

tential metabolic oxidation sites (16, 17) proved to be fruitful

in generating stable analogues.

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 20

&

7

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

Metabolic identification studies in rat hepatocytes showed

that 9 is extensively metabolized. The major sites of oxidative

metabolism were shown to be the pendent cyclopentyl ring as

well as the ethyl group, with demethylation accounting for

only a small fraction of the total metabolites identified; de-

pending on substitution, other analogues could suffer from

metabolism on the biaryl group as well. A two-pronged strat-

egy was employed to address this problem. Efforts were un-

dertaken to overcome this undesired oxidative metabolism by

lowering the logP of subsequent analogues and through di-

rected heteroatom substitution. In the context of the bicyclic

pteridine core, these focused efforts produced advanced ana-

logues, one of which was used to generate promising evidence

of target engagement in vivo.^[19]However, our experience with

methodical derivatization at the 8-position of the pteridinone

ring compelled a more radical approach to variation of this

sector. This was attempted by surveying analogues that kept

the same basic architecture as 9, but incorporated a fused cy-

cloalkyl ring on the pteridinone core to provide 7,8-tricyclic

pteridinone analogues (Figure 4).

Figure 4. Increasing oxidative metabolic stability through cyclization and

lowered logP.

fold decrease in Plk-2 potency, there was also roughly a 100-

fold improvement in Plk-2/Plk-1 selectivity of the 7,8-tricyclic

pteridinone analogues relative to 9. The exact reason for this

has remained elusive. One possibility is that the tricyclic com-

pounds undergo a more optimal edge-to-face interaction with

Tyr161 of the hinge. An alternative explanation rests on the

subtle changes in protein conformation that modulate the in-

teraction of the P-loop with the inhibitors. Comparisons be-

tween published Plk-1 and our Plk-2 crystal structures indicate

that Plk-2 has a wider gap between the N and C domains of

the kinase. The shift in the N and C domains caused a 1–2 ꢁ

shift in the P-loop in the vicinity of the 8-substituent on the

pteridinone ring. However, the pursuit of more detailed struc-

Gratifyingly, lowering logP by fusing a ring to the pteridi-

none core did indeed improve stability toward oxidative me-

tabolism, as shown in Table 3. Despite the approximate three-

Table 3. Lowering logP improves oxidative metabolic stability in 7,8-tricyclic pteridinones.

Compound

IC₅₀[mm]^[a]

Plk-1

Selectivity

Plk-2/Plk-1

EC₅₀[mm]

HEK293

logP

OxMet

Permeability

Solubility

[b]

Plk-2

Plk-3

–

(m,r,h)^[b]

[nms^ꢀ1

]

[mm]^[b]

18

19

0.166

>10

60

–

1.9

76, 66, 88

50, 75, 72

243

100

0.063

0.017

7.2

30.750

9.676

114

4.48

2.3

3.3

249

190

20

21

22

4.22

27.3

55.9

248

700

1.10

1.20

4.70

19, 2, 6

41, 60, 69

50, 34, 49

35

11

6

0.039

0.053

6.445

2.3

128

42.863

1054

–

[a] Values are means of at least two experiments at 100 mm ATP concentration; details can be found in the Supporting Information. [b] See ref. [23].

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 20

&

8

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

tural analyses would have required a direct comparison of Plk-

2 and Plk-1 complexes with these compounds, and generation

of the latter was outside of the scope of this effort.

ty, and permeability (Table 4). Pyridines as A-ring replacements

provided little advantage over the imidazole found in 21. De-

spite its favorable P-glycoprotein (P-gp) ratio, 23 fell short of

the level of potency, Plk-2/Plk-1 selectivity, and oxidative meta-

bolic stability displayed by 21. Compound 24 exhibited poten-

cy similar to 21, but it suffered from poorer selectivity and oxi-

dative stability compared with 21. Despite its low logP and ex-

cellent oxidative metabolic stability, triazole 25 showed a dra-

matic 250-fold loss in potency relative to 21 and a complete

loss of Plk-2/Plk-1 selectivity. Indeed, the increased electron

density proximal to Tyr161 may have induced a Coulombic re-

pulsion, as evidenced by lower potency and selectivity. The iso-

xazole A-ring replacement 26 was also found to be suboptimal

in all of the desired properties. Thiadiazole 27 was the most

promising of the A-ring replacements that were examined, as

it provided the greatest degree of Plk-2/Plk-1 selectivity, with

potency and oxidative stability on par with that of 21. Unfortu-

nately, 27 proved insoluble under standard conditions. None

of these analogues proved to be comparably potent in cell or

enzyme activity to the previously synthesized inhibitors listed

in Table 2.

Increasing ring size from the pyrrolidine 18 to the six-mem-

bered piperidine 20, morpholine 21, or piperazine 22 im-

proved Plk-2 activity. Compound 19 illustrates the importance

of the ethyl group for potency, as is evident by the 2.6-fold in-

crease in potency relative to 18. Despite its excellent potency

and Plk-2/Plk-1 selectivity, with a measured logP=3.3, com-

pound 20 demonstrated only modest oxidative metabolic sta-

bility relative to 18 and 19. Given the measured logP for 20, it

was not surprising to observe a decrease in aqueous solubility

relative to 19. Surprisingly, a dramatic decrease in solubility for

21 was observed, despite its lower measured logP, concomi-

tant with a loss in permeability. Poor solubility notwithstand-

ing, 21 offered the best combination of potency, oxidative

metabolic stability, and Plk-2/Plk-1 and Plk-2/Plk-3 selectivity of

the group. Compound 21 was evaluated in the kinase selectivi-

ty panel featuring 35 kinases (Supporting Information table 1)

and an expanded selectivity panel of 303 kinases (Supporting

Information table 2). It demonstrated excellent selectivity in

both kinase panels: only MAP4K4 was inhibited at >80% at

10 mm concentration. The N-methyl replacement analogue 22,

compared with 21, showed a modest loss in activity, both in-

trinsic and cellular, a modest loss in oxidative metabolic stabili-

ty and solubility, but a 1.5-fold gain in Plk-2/Plk-1 selectivity.

Evaluation of the more potent analogues 19–22 in the cellu-

lar model showed that despite the lower measured logP

values of these compounds, the cellular activity remained two

orders of magnitude lower than the enzymatic activity. This

was true despite excellent in vitro passive permeability

throughout the series. Nevertheless, compound 21 emerged as

the leading candidate with the best overall properties for fur-

ther development.

The series of B-ring replacements brought mixed results

with respect to the desired improvements over 21 in potency,

Plk-2/Plk-1 selectivity, and physicochemical properties (Table 5).

Compound 28 realized slight improvements in microsomal sta-

bility, solubility, and Plk-2/Plk-1 selectivity, but a slight loss in

potency relative to 21. Compound 29 displayed a twofold in-

crease in potency over 21, but a modest loss in Plk-2/Plk-1 se-

lectivity. The B-ring of 30 was substituted with 4-fluoropyridyl-

phenyl in an attempt to increase solubility. Despite the gain in

microsomal stability, the solubility of 30 was the same as that

of 21, and the potency and Plk-2/Plk-1 selectivity were de-

creased by 9.5- and 2.5-fold, respectively. A similar increase in

microsomal stability was observed for 31, in addition to favora-

ble solubility and permeability. However, the decrease in po-

tency and selectivity were more pronounced, at 22.5- and 6-

fold, respectively.

To capitalize on the excellent kinome selectivity, potency,

and oxidative stability demonstrated by 21, a focused group of

analogues with this advanced core was made by targeting A-

and B-ring modifications, as shown in Figure 5. A-ring modifi-

In vivo pharmacokinetic studies

Compounds 3, 17, and 21 were unique in these series because

of their stability in the liver microsome assays, permeability,

and potency against Plk-2. Each compound was examined in

an in vivo model of brain exposure in mouse, as well as by ob-

taining the i.v./p.o. pharmacokinetic profiles in rat (Table 6). In

the mouse brain uptake studies the rodents were given

a 1 mgkg^ꢀ1i.v. bolus dose of either 3, 17, or 21, and brain

tissue and plasma were sampled at each time point. The re-

sults are reported as the AUC over a 3 h time course in both

brain and plasma compartments and also expressed as a ratio

of AUC values to give the brain to plasma (B/P) ratio. Com-

pound 3 demonstrated excellent brain exposure in mouse; 21

showed lower but significant exposure, while 17 exhibited no

significant brain exposure in mouse. The low brain exposure is

consistent with the measured permeability for 17. The efflux

ratio both with and without the P-gp transport inhibitor tari-

Figure 5. Modifications to the 7,8-tricyclic morpholine pteridines.

cations were performed in an effort to increase Plk-2/Plk-1 se-

lectivity, whereas B-ring modifications were explored in an

effort to improve on the potency and physicochemical proper-

ties displayed by 21.

The A-ring replacements were examined in an attempt to in-

crease Plk-2/Plk-1 selectivity, as well as to improve certain

physicochemical properties such as solubility, oxidative stabili-

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 20

&

9

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

Table 4. A-ring modifications affect Plk-2/Plk-1 selectivity.

Compound

IC₅₀[mm]^[a]

Plk-1

Selectivity

Plk-2/Plk-1

EC₅₀[mm]

HEK293

logP

OxMet

(m,r,h)^[b]

Permeability

Solubility

[mm]^[b]

[b]

Plk-2

Plk-3

[nms^ꢀ1

]

23

24

0.104

7.23

17.6

2.591

70

1.90

8.20

3.0

0, 25, 34

0, 37, 66

175

25

65

0.059

9.91

2.458

4.006

298

2.3

1.6

155

91

25

26

27

11.1

18.2

1

95

–

74, 87, 88

58, 6, 77

45, 11, 52

38

35

0

0.192

0.057

12.506

4.966

2.0

50

>100

1770

>10

–

126

[a] Values are means of at least two experiments at 100 mm ATP concentration; details can be found in the Supporting Information. [b] See ref. [23].

Table 5. B-ring substituent effects on potency and pharmacokinetics.

Compound

IC₅₀[mm]^[a]

Plk-1

Selectivity

Plk-2/Plk-1

EC₅₀[mm]

HEK293

logP

OxMet

Permeability

Solubility

[b]

Plk-2

0.056

Plk-3

(m,r,h)^[b]

[nms^ꢀ1

]

[mm]^[b]

28

29

42.9

8.71

20.648

766

484

3.30

1.60

2.6

47, 67, 81

19, 33, 46

169

35

25

0.018

1.874

–

88

30

31

0.371

0.879

>100

53.022

270

114

–

3.3

1.6

59, 86, 91

77, 92, 96

164

213

11

>100

100

[a] Values are means of at least two experiments at 100 mm ATP concentration; details can be found in the Supporting Information. [b] See ref. [23].

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 20

&10

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

(Figure 6a). The levels of total a-synuclein were unaf-

fected by this treatment (Figure 6b). The brain-to-

plasma ratios derived from the exposure data for 21

at 6 h following oral dosing in rats (Table 7) are in

close agreement with the ratio predicted by the

mouse brain uptake model (Table 6). The 6 h brain

concentrations for 21 at the 100 and 200 mgkg^ꢀ1

doses were respectively five- and sevenfold greater

than the Plk-2 cellular IC₅₀value. As expected in vehi-

cle-treated rats, total a-synuclein reached 60–

70 mgg^ꢀ1with phospho-S129 a-synuclein represent-

ing only 0.1% of the total. In conclusion, compound

21 proved useful as a probe of Plk-2 in the CNS com-

partment, because it is not a P-gp substrate, and ex-

hibits good permeability and high oral bioavailability.

This compound was well tolerated in the target en-

gagement study, and sustained 6.8 mm levels in the

Table 6. In vivo PK parameters.

Compound:

3

17

21

Mouse brain uptake, 1 mgkg^ꢀ1i.v.

plasma AUC_(0–3h)[ngh^ꢀ1mL^ꢀ1

]

205

346

1.7

2626

88

0.03

660

223

0.3

brain AUC_(0–3h)[ngh^ꢀ1g^ꢀ1

brain/plasma AUC ratio

]

Rat PK, 1 mgkg^ꢀ1i.v.

AUC_(0–1)[ngh^ꢀ1mL^ꢀ1

]

123ꢁ10

0.4ꢁ0.2

8191ꢁ727

247

0.4

3326ꢁ128

545

543ꢁ82

0.5

1871ꢁ298

47

0.5ꢁ0.1

922ꢁ344

151

449ꢁ64

1.5ꢁ0.5

2256ꢁ297

57

0.5ꢁ0.1

1164ꢁ84

191

t_1/2[h]

CL [mLh^ꢀ1kg^ꢀ1

Q_hepatic[%]

MRT [h]

]

V_dss[mLkg^ꢀ1

]

BW_total[%]

Rat PK, p.o.

dose [mgkg^ꢀ1

]

2

10

2

C_max[ngmL^ꢀ1

t_max[h]

]

60ꢁ73

0.3ꢁ0.1

32ꢁ35

0.3ꢁ0.1

0.6ꢁ0.2

13

7593ꢁ1696

0.25

3729ꢁ1029

1.76ꢁ0.15

0.5

1613ꢁ478

0.25

1310ꢁ916

0.8ꢁ0.29

0.7ꢁ0.2

153

CNS up to 6 h after oral dosing in rat at 100 mgkg^ꢀ1

.

AUC_(0–1)[ngh^ꢀ1mL^ꢀ1

t_1/2[h]

]

MRT [h]

F [%]

69

Chemistry

The general synthetic route for the preparation of

the various pteridinones listed in Tables 1 and 2 is il-

lustrated in Scheme 1. Chloropteridinone intermedi-

quidar present (P_eff(AꢀB)/P_eff(BꢀA)=27) indicates that 17 is a mod-

erate substrate for this transporter.

The rat i.v./p.o. pharmacokinetic profile for 3, 17, and 21 was

reasonably consistent with the trends suggested by the in vitro

ADME data. The poor microsomal stability in rat for 3 (16% of

3 remaining after a 30 min incubation) in addition to a possible

extrahepatic metabolic pathway translated into large values for

i.v. clearance (Table 6). Both 17 and 21 exhibited good stability

in rat microsomes. Although the rates of clearance remained

high, both 17 and 21 showed moderately improved rates of

both absolute clearance and clearance relative to hepatic

blood flow. All three compounds demonstrated excellent per-

meability in the in vitro MDCKII MDR-1 model. Good plasma

exposures after oral administration and a high fraction of com-

pound absorbed were observed for both 17 and 21. However,

the peripheral plasma exposures displayed by compound 17

were substantial. A 10 mgkg^ꢀ1single oral dose yielded an AUC

of 3730 ngh^ꢀ1mL^ꢀ1with an oral t_1/2of 1.76 h (Table 6), making

17 a suitable probe of the role of Plk-2 in the peripheral com-

partment rather than the CNS. The same parameters were con-

siderably lower for 3, consistent with the poor microsome sta-

bility and high i.v. clearance in rat. Based on the observed

brain exposure and PK profile, 21 was advanced to an in vivo

target engagement study in rat.

Target engagement

A CNS target engagement study confirmed the ability of 21 to

cross the blood–brain barrier and achieve brain levels sufficient

to inhibit the endogenous Plk-2-mediated phosphorylation of

a-synuclein in rat brain. Administration of a single oral 100 or

200 mgkg^ꢀ1dose of 21 to Sprague–Dawley rats resulted in re-

spective 34.5 and 41.4% decreases (p<0.0001) in phospho-

S129 a-synuclein levels in the cerebral cortex for up to 6 h

Figure 6. A single oral dose of 21 a) significantly decreases phospho-S129 a-

synuclein levels in rat cerebral cortex, whereas b) total a-synuclein levels are

unchanged. Data represent the mean ꢁSD with n=6 for each group; the

inhibition for each dose reaches significance (***p<0.0001) versus vehicle

control group (Veh) as determined by one-way ANOVA followed by Tukey’s

post hoc test.

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 20

&

11

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

from l-proline.^[28]Reaction of amino ester 36 with 2,4-dichloro-

5-nitropyrmidine (37) exclusively provided regioisomer 38. In-

tramolecular iron-mediated reductive cyclization followed by

methylation with trimethyl phosphate provided 39, which

could then be treated with 2-phenylimidazole to yield 19.

In an analogous manner, the coupling of amine 40^[29]with

excess 2,4-dichloropyrimidine gave an 8:1 mixture of the de-

sired 4-substituted product 41 (67% yield) and the regioiso-

meric 2-substituted byproduct. Compound 41 was transformed

into key intermediate 42 before displacement and chiral purifi-

cation to afford final product 21.

Table 7. Mean plasma (P) and brain (B) levels of 21 at 6 h following

a single oral dose.

Dose [mgkg^ꢀ1

]

Time [h]

P [ngmL^ꢀ1

]

B [ngg^ꢀ1

]

B/P ratio

100

200

6

7756ꢁ2032

8141ꢁ3556

2629ꢁ1612

3380ꢁ1331

0.46

0.42

Advanced intermediate 42 provided a good point for diver-

sification in the production of A-ring modifications, as shown

in Scheme 3. Advanced intermediate 42 was partitioned down

three divergent reaction paths. Suzuki reaction of 42 with the

appropriate boronic acid (e.g., 43) provided analogues 23–24

(Table 2). Sonogashira coupling of 42 with phenylacetylene fol-

lowed by reaction with sodium azide afforded triazole 25. The

palladium-catalyzed reaction of 42 with acetophenone provid-

Scheme 1. General synthetic strategy. Reagents and conditions: a) N-hetero-

cycle, D, Cs₂CO₃, DMF or DMSO; b) Ullmann coupling; c) Suzuki coupling.

ates 32–35 were prepared ac-

cording to published proce-

dures.^[20]These intermediates

could undergo heating with the

appropriate N-heterocycle to

provide the desired compounds.

Alternatively, they could first be

subjected to Ullmann or Suzuki

coupling^[27]to provide the de-

sired compounds. The detailed

syntheses of all compounds

shown are described fully in the

Supporting Information.

A representative synthesis of

these 7,8-tricyclic pteridinone

analogues is shown in Scheme 2.

Each synthesis proceeded in

a straightforward manner from

Scheme 3. A-ring modifications. Reagents and conditions: a) phenylacetylene, Pd(PPh₃)₄, CuI, Et₃N, CH₃CN, MW,

1408C, 5 min; b) NaN₃, DMSO, MW, 1208C, 1 h; c) Na₂CO₃, Pd(dppf)Cl₂, DME/H₂O, 908C; d) acetophenone, (1,3-

bis(2,6-diisopropylphenyl)-4,5-dihydroimidazol-2-ylidene)Pd(Cl)allyl, tBuONa, THF, reflux; e) N₂H₄, EtOH, reflux,

f) SOCl₂; g) DMF/DMA, DMF, AcOH, NH₂OH·HCl; h) chiral HPLC.

the appropriate starting amino

ester. The synthesis of 19 re-

quired the enantiopure amino

ester 36, derived in four steps

Scheme 2. Representative synthesis of the 7,8-tricyclic pteridinone analogues. Reagents and conditions: a) DIPEA, THF, 08C; b) Fe, AcOH, 908C; c) (CH₃)₃PO₄,

K₂CO₃, dioxane, 908C; d) 2-phenylimidazole, Na₂CO₃, DMF, 1508C; e) chiral HPLC.

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 20

&12

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

Data collection and structure determination: Diffraction data were

collected on beamline 5.0.2 at ALS and were processed using HKL-

2000.^[33]The first Plk-2 structure was determined by molecular re-

placement using Plk-1 (PDB code 2RKU) as a searching model with

the ligand and water molecules removed. The subsequent rounds

of manual model building and TLS refinement were performed

using Coot^[34]and Refmac 5, respectively.^[35]PDB entry designations

for Plk-2 complexes with compounds 1, 1a, 4, 9, and 11 are shown

in the legends of Figures 1 and 2.

ed advanced intermediate 44, which could either be treated

with DMF/DMA and hydroxylamine to afford 26, or with hydra-

zine and thionyl chloride to give 27.

Conclusions

These efforts resulted in novel, potent, and selective inhibitors

of Plk-2, with pharmaceutical properties sufficient to enable

successful in vivo target-engagement studies with compound

21. The now commonly used structure-guided discovery ap-

proach to engineer a ligand-efficient core for increased poten-

cy and selectivity by crystallographic and computational

means provided a powerful framework for these compounds

to be designed and synthesized in an efficient manner. By al-

tering the design of the hinge region interface of the mole-

cules to use a tandem hydrogen bond/aromatic edge-to-face

interaction and effectively exploiting the space between

Tyr161 and Arg165, the overall selectivity of these analogues

versus non-Plk kinases was significantly enhanced. Incorporat-

ing a fused ring into the pteridine core further enhanced both

selectivity and pharmaceutical properties, and gave rise to key

compound 21.

Compound 21 is a useful tool for understanding the conse-

quence of inhibiting Plk-2 on synuclein phosphorylation in rat

brain. Continuing focus is placed on confirming the hypothesis

that a-synuclein phosphorylation is a key step in the patho-

physiology of Parkinson-like symptoms.^[30]The proof-of-princi-

ple studies in a rodent model of transgenic^[31]or virally induced

synuclein overexpression^[32]will require a well-tolerated agent

capable of suppressing synuclein phosphorylation in the CNS

compartment over a sustained period of time. Compound 21

was identified as a compound that is highly selective for the

target and achieved the concentrations required in vivo to ef-

fectively inhibit synuclein phosphorylation by Plk-2 in wild-

type rat brain. Further studies are required to define the effect

of repeated oral dosing on a-synuclein phosphorylation in

rodent brain.

Molecular dynamics simulations

Compound synthesis was guided by computer simulation. Each

candidate for chemical synthesis was evaluated for its potential

binding affinity to the protein using the molecular mechanics Pois-

son–Boltzmann surface area (MM-PBSA) methodology as imple-

mented in the AMBER 10 molecular simulation program suite.^[36]

Compounds with higher predicted affinities were prioritized for

synthesis.

The MM-PBSA method calculates protein–ligand binding affinity by

taking the average interaction and solvation energies over a set of

conformations obtained from a room-temperature molecular dy-

namics (MD) simulation of the solvated protein–ligand complex.

The protocol to obtain this set of complex structures is as follows:

The ligand to be evaluated was constructed by in situ modification

of protein–ligand complex (obtained from X-ray crystallography)

using the Maestro molecular modeling package (Schrçdinger, Inc.).

MD parameters for the ligand were obtained from the ANTECHAM-

BER module.^[37]AMBER’s solvateCap feature was used to add water

molecules to a sphere of 29 ꢁ radius centered on the ligand. Pro-

tein atoms on residues with a-carbons 15 ꢁ beyond all ligands

atom were frozen during the simulation.

This initial structure was equilibrated to a room-temperature pro-

duction MD run in eight sequential steps (Table 8). The conforma-

tions of the solvated protein–ligand complex were sampled at 1 ps

intervals in the production MD run (Step 8 in Table 8). Binding free

energies were computed for each sampled conformation and aver-

aged over the 250 samples to obtain the estimated affinity for the

ligand. Replicate simulations were conducted by using slightly

varied starting coordinates (1% of coordinates randomly perturbed

by ꢁ0.001 ꢁ). Eight replicate simulations using the protocol in

Table 8 were then averaged to obtain the final estimate of ligand

binding energy.

Experimental Section

Selectivity panel of single-point kinase inhibition data

X-ray crystallography

The kinase profiling panel used, SelectScreen, is provided as a com-

mercial service by InVitrogen/Life Technologies (Grand Island, NY,

USA). Compounds were tested at a concentration of 10 mm, while

the ATP concentration used in these assays was 2ꢂK_M; the assays

are a mixture of formats: Z’-LYTE, Adapta, and LanthaScreen tech-

nologies.

Crystallization: Plk-2 kinase domain A2 mutant protein at

8 mgmL^ꢀ1in buffer (final buffer) was incubated on ice with 1 mm

staurosporine for 30 min. The drops were set up with a 1:1 (v/v)

ratio of protein to mother liquor in a total volume of 2 mL. Diffrac-

tion-quality crystals of Plk-2 kinase domain in complex with the

compound were obtained by the sitting-drop vapor diffusion

method at 277 K against a reservoir containing 0.1m Bis-Tris

pH 5.5, 0.8m sodium formate, 19% PEG 3350. Crystals appeared in

2–3 days and developed to full size in 2–3 weeks. Crystals were

mounted directly from the drops and flash-frozen in liquid nitro-

gen. Apo-crystals can be obtained under the same conditions de-

scribed above, except that a low-affinity small fragment compound

was added at 1 mm concentration in an attempt to obtain co-crys-

tals with this compound. The crystal structure showed the com-

pound to be absent.

In vivo studies

All animal use procedures were approved by the Institutional

Animal Care and Use Committee (IACUC) at Elan Pharmaceuticals

and performed in accordance with the National Research Council

Guide for the Care and Use of Laboratory Animals (Institute of Lab-

oratory Animal Resources, National Research Council, 1996), and all

applicable regulations.

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 20

&13

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

Table 8. MD protocol for the MM/PBSA estimation of binding energies.

Step

1

Description

Convergence Criterion

or Duration^[a]

Constrained Atoms

Magnitude of

T [K]^[b]

Restraint [kcalmol^ꢀ1ꢁ^ꢀ1

]

Minimization: water H atoms

0.01 RMSD

All protein and ligand atoms

and water molecule oxygen atoms

All protein and ligand atoms

All protein atoms

Protein a-carbon atoms

10.0

–

2

3

4

5

6

7

8

Minimization: water molecules

Dynamics: water molecules

Minimization: ligand and water molecules

Minimization: entire system

Dynamics: raise temperature

Dynamics: equilibration

0.01 RMSD

25 ps

0.01 RMSD

50 ps

4.0

1.5

–

0–200

–

0–300

300

100 ps

250 ps

Dynamics: production run

[a] For minimization steps, the RMSD convergence criterion is listed; for dynamics steps, the time of the simulation is listed. [b] For temperature ranges,

the temperature was linearly scaled over the time interval of the simulation.

est. Data acquisition and analysis was conducted using Analyst ver-

sion 1.5.1.

Pharmacokinetics assays

Two- to three-month old female FVB mice were obtained from Ta-

conic Farms (Hudson, NY, USA), and male jugular-vein-cannulated

Sprague–Dawley rats (approximately 8 weeks of age) were ob-

tained from Charles River Laboratories (Hollister, CA, USA).

Pharmacokinetic analysis: For mice mean composite plasma con-

centration–time and brain concentration–time profiles were gener-

ated for each test article. Non-compartmental pharmacokinetic

analysis using the composite profiles was conducted using Win-

Nonlin Professional Version 5.2 (Pharsight, Mountain View, CA,

USA). Individual rat plasma profiles following i.v. and p.o. dosing

were used to calculate pharmacokinetic parameters. Parameters

determined include, as applicable, area under the plasma concen-

tration–time profile (AUC_(0–1)), the terminal half-life (t_1/2), clearance

(CL), volume of distribution at steady-state (V_dss), time to maximum

concentration (t_max), and maximum concentration (C_max). The ratio

of brain concentration to plasma concentration (B/P ratio) was cal-

culated using AUC_(0–1)values and individual time points. Oral bio-

availability was calculated at the ratio of the dose-normalized

AUC_(0–)following p.o. dosing and AUC_(0–)following i.v. dosing. Sum-

mary statistics were calculated using MS Excel, MS Office 2003

(Redmond, WA, USA).

Mouse brain uptake assay: Test compounds were formulated for i.v.

dosing in a vehicle containing 5% Solutol/10% propylene glycol in

saline on the day of dosing. Compounds were formulated at a con-

centration of 0.2 mgmL^ꢀ1and administered via tail vein bolus at

a dosage of 1 mgkg^ꢀ1. Samples were collected (three mice per

time point) at 0.083, 0.5, 1, 3, and 6 h post-dose. Blood samples,

collected by cardiac puncture, were transferred to tubes containing

tripotassium EDTA and stored on ice prior to centrifugation. Sam-

ples were centrifuged at 10000 rpm at 48C for 5 min to obtain

plasma. Dissected brain samples (right hemisphere) were rinsed,

weighed, and immediately frozen with dry ice. All samples were

stored at ꢀ808C prior to bioanalysis.

Rat pharmacokinetic assay: Rats received either a 1 mgkg^ꢀ1intrave-

nous bolus (n=3) or a 2 mgkg^ꢀ1oral gavage dose of test article.

Test articles for intravenous (i.v.) administration were formulated as

solutions in Solutol/propylene glycol/water (10:20:70% v/v/v) for

sterile injection. The dose volume for i.v. administration was

1 mLkg^ꢀ1. Oral doses were formulated as suspensions in 0.5%

Target engagement study: inhibition of Plk-2 phosphorylation

of a-synuclein in rat brain

Dosing and tissue collection: Male Sprague–Dawley rats (Charles

River, Gilroy, CA, USA) weighing ~100–120 g were given compound

Tween 80 in 2% methylcellulose at a dose volume of 5 mLkg^ꢀ1

.

Animals receiving oral doses were fasted overnight and for 4 h

after dosing. Following test article administration, serial blood sam-

ples at 0.083 (i.v. only), 0.25, 0.5, 1, 2, 4, 6, 8, 12, and 24 h into

K₃EDTA vials. Samples were transferred to tubes containing tripo-

tassium EDTA and stored on ice prior to centrifugation. Samples

were centrifuged at 10000 rpm at 48C for 5 min to obtain plasma.

All samples were stored at ꢀ808C prior to bioanalysis.

20 orally (in 5% DMSO in 1% methylcellulose, volume: 10 mLkg^ꢀ1

)

as a single dose of either 100 or 200 mgkg^ꢀ1(n=6 per group).

Rats were euthanized by CO₂narcosis 6 h after dosing, and brains

were removed quickly; the cerebral cortex was dissected out,

placed on ice, and frozen at ꢀ808C until quantification of total

and phosphorylated a-synuclein by ELISA.

Quantification of a-synuclein levels by ELISA: Because Plk-2 has

been proposed as the main enzyme responsible for a-synuclein

phosphorylation at serine 129,^[3,4]we monitor the changes in phos-

pho-S129 a-synuclein levels as a measure of Plk-2 activity in the

brain. The total a-synuclein and phospho-S129 a-synuclein levels

were quantified with a sandwich ELISA as described previously^[2c,3]

with some modifications.

Bioanalysis: Prior to analysis, brain tissue was homogenized with

five parts 50 mm ammonium acetate buffer (pH 7.4). Test article

concentrations in rat and mouse plasma and mouse brain homo-

genate samples were analyzed by using an LC–MS–MS method.

Samples were prepared by protein precipitation using CH₃CN (with

internal standard). Following mixing and centrifugation the super-

natant was injected into the LC–MS–MS system. The LC–MS–MS

system was composed of an Agilent 1290 LC HPLC system (Santa

Clara, CA, USA), a CTC HTC Autosampler (Leap Technologies, Carr-

borro, NC, USA) and an ACE 2.1ꢂ50 mm C₁₈1.8 mm LC column

(MacMOD Analytical Inc. Chadds Ford, PA, USA). Separation was

Brain samples were homogenized with a bead-based Qiagen Tis-

suelyzer in 0.4% CHAPS solution containing phosphatase and pro-

tease inhibitors. The resulting homogenate was centrifuged at

14000 g for 20 min in an Eppendorf microcentrifuge. The superna-

tant CHAPS-soluble fraction was removed and diluted 1:20 with

cell extraction buffer consisting of 0.1% SDS, 0.5% deoxycholate,

achieved using a 2.5 min gradient at a flow rate of 0.6 mLmin^ꢀ1

Multi-reaction monitoring was used to detect the analytes of inter-

.

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 20

&14

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

1% Triton X-100 and phosphatase inhibitors at a final concentra-

tion of 20%. For the detection of total a-synuclein, the monoclonal

antibody 1H7 (amino acids 91–99) was used as the capture anti-

body and the biotinylated 5C12 antibody (amino acids 109–120) as

the reporter antibody. For the detection of phospho-S129 a-synu-

clein levels, 1H7 was used as the capture antibody and biotinylated

11A5 (amino acids 124–134, specific to S129 phosphorylation) as

the reporter antibody. The detection limit for phospho-S129 a-syn-

Reversed-phase HPLC methods

Analytical method A: The initial solvent composition was 20%

CH₃CN with 0.1% trifluoroacetic acid (TFA) and H₂O with 0.1% TFA

which ramped to 70% CH₃CN over 10 min, held at 70% for 2 min,

then ramped to 95% over 1 min, and held at 95% for 2 min with

a flow rate of 1.5 mLmin^ꢀ1at 358C.

Analytical method B: The same parameters as Method A changed

so that the initial solvent composition was 50% CH₃CN which

uclein was 0.005 mgg^ꢀ1

.

ramped to 95% CH₃CN over 10 min with

1.5 mLmin^ꢀ1, at 358C.

a flow rate of

Representative synthetic procedures and exemplified ana-

logues

Analytical method C: The same parameters as Method A changed

so that the initial solvent composition was 20% CH₃CN which

ramped to 50% CH₃CN over 10 min with

a flow rate of

Reagents and solvents obtained from commercial suppliers were

used without further purification unless otherwise stated. Imida-

zole precursors for inhibitors 28–31 were synthesized by published

methods.^[38]Thin-layer chromatography was performed on pre-

coated 0.25 mm silica gel plates (E. Merck, silica gel 60 F₂₅₄). Visuali-

zation was achieved using UV illumination or by staining with

phosphomolybdic acid, ninhydrin, or other common staining re-

agents. Flash chromatography was performed with either a Biotage

Flash 40 system and pre-packed silica gel columns or hand-packed

columns (E. Merck silica gel 60, 230–400 mesh). Preparatory HPLC

was performed on a Varian Prepstar high-performance liquid chro-

1.5 mLmin^ꢀ1, at 358C.

Analytical method D: The same parameters as Method A changed

so that the initial solvent composition was 5% CH₃CN which

ramped to 20% CH₃CN over 10 min with

1.5 mLmin^ꢀ1, at 358C.

a flow rate of

Analytical method E: Solvent A: H₂O (0.05% TFA), solvent B:

CH₃CN (0.05% TFA); gradient: 5!95% B in 1.4 min; flow rate:

2.3 mLmin^ꢀ1; column: SunFire C₁₈, 4.6ꢂ50 mm, 3.5 mm; oven tem-

perature: 508C.

1

matograph. All final compounds were obtained in ꢂ95% purity. H

Analytical Method F: Mobile phase A: H₂O (0.01% NH₃), phase B:

and ¹³C NMR spectra were recorded at 300, 400, or 500 MHz and

75, 100 or 125 MHz, respectively, on a Varian Gemini or Bruker

Avance spectrometer. Chemical shifts (d) are reported in parts per

million downfield relative to tetramethylsilane or to proton reso-

nances resulting from incomplete deuteration of the NMR solvent.

Mass spectra were recorded on an Agilent series 1100 mass spec-

trometer connected to an Agilent series 1100 HPLC. Elemental

analyses were performed at ALS Environmental (Tucson, AZ, USA).

CH₃CN; gradient: 5!95% B in 1.6 min; flow rate: 1.8 mLmin^ꢀ1

;

column: XBridge C₁₈, 4.6ꢂ50 mm, 3.5 mm; oven temperature: 408C.

(R)-8-Cyclopentyl-7-ethyl-5-methyl-7,8-dihydropteridin-6(5H)-

1

one (1a): H NMR (500 MHz, CDCl₃): d=8.28 (s, 1H), 7.79 (s, 1H),

4.45–4.35 (m, 1H), 4.17 (dd, J=7.8, 3.8 Hz, 1H), 3.27 (s, 3H), 2.05–

1.95 (m, 1H), 1.95–1.85 (m, 1H), 1.85–1.70 (m, 4H), 1.70–1.50 (m,

3H), 0.78 ppm (t, J=7.5 Hz, 3H); LC–MS m/z: 261.0 [M+H]⁺; t_R=

1.37 min (Method E); HRMS (ESI+) m/z calcd for C₁₄H₂₁N₄O⁺

261.1710, found 261.1710.

LC–MS and HRMS analysis on a Q-TOFMS: Stock solutions

(20 mm) for all standard compounds prepared in DMSO were dilut-

ed in water to 5 mm for positive-mode mass spectrometric analysis.

This typically produced maximum ion abundances on the order of

1ꢂ10⁶–5ꢂ10⁶ions per scan, although for data processing, spectra

with ion abundances of 5ꢂ10⁵ions/scan were averaged. A triple-

TOF 5600 quadrupole time-of-flight hybrid mass spectrometer (AB

Sciex, Foster City, CA, USA) equipped with an LC system consisting

of a LEAP PAL HTC autosampler and an Agilent 1290 Infinity LC

pump (Agilent, Santa Clara, CA, USA), was used for LC–MS analysis.

The mass spectrometer was operated in the positive ionization

mode with a spray voltage of 5.5 kV. Ion source gas 1, ion gas

source 2, and curtain gas were set at 40, 45, 250 psi, respectively;

the source temperature was maintained at 5008C. Samples (5 mL)

were introduced to the mass spectrometer via the HPLC system

using an Agilent EC-C₁₈column (2.1ꢂ75 mm, 2.7 mm) under iso-

cratic conditions (50:50 CH₃CN/H₂O with 0.1% formic acid) at

(R)-8-Cyclopentyl-7-ethyl-5-methyl-2-(pyrrolidin-1-yl)-7,8-dihy-

dropteridin-6(5H)-one (2): ¹H NMR (400 MHz, CDCl₃): d=7.61 (s,

1H), 4.20 (t, J=8.0 Hz, 1H), 4.15 (q, J=3.6 Hz, 1H), 3.60–3.48 (m,

4H), 3.28 (s, 3H), 2.20–1.90 (m, 7H), 1.90–1.75 (m, 4H), 1.75–1.55

(m, 3H), 0.86 ppm (t, J=7.6 Hz, 3H); LC–MS m/z: 330.0 [M+H]⁺;

t_R=1.52 min (Method E); HRMS (ESI+) m/z calcd for C₁₈H₂₈N₅O⁺

330.2288, found 330.2291.

(R)-8-Cyclopentyl-7-ethyl-2-(1H-imidazol-1-yl)-5-methyl-7,8-dihy-

dropteridin-6(5H)-one (3): ¹H NMR (500 MHz, CDCl₃): d=8.50 (s,

1H), 7.78 (s, 1H), 7.76 (s, 1H), 7.14 (s, 1H), 4.33–4.26 (m, 2H), 3.37

(s, 3H), 2.17–2.10 (m, 1H), 2.05–1.82 (m, 6H), 1.80–1.67 (m, 3H),

0.88 ppm (t, J=7.5 Hz, 3H); LC–MS (0.01% NH₃) m/z: 327.2 [M+

H]⁺; t_R=1.97 min (Method F); Anal. calcd for C₁₇H₂₂N₆O: C 62.56, H

6.79, N 25.75, found C 62.38, H 6.91, N 25.55.

a flow rate of 0.8 mLmin^ꢀ1

.

(R)-8-Cyclopentyl-7-ethyl-5-methyl-2-(1H-pyrrol-2-yl)-7,8-dihy-

1

dropteridin-6(5H)-one (4): H NMR (500 MHz, CDCl₃): d=9.66 (brs,

Full-scan mass spectra were acquired over the range of 120–

900 m/z with an accumulation time of 0.25 s. Total acquisition time

was 2 min per sample, and all four time-to-digital converter chan-

nels were used for detection. The tripleTOF instrument was cali-

brated by an integrated calibration delivery system (AB Sciex)

using the manufacturer’s positive calibration solution, at an injec-

tion flow rate of 300 mLmin^ꢀ1. These calibrant ions were intro-

duced via an orthogonal atmospheric pressure chemical ionization

source in the presence of mobile phase flow. Mass spectral data ac-

quisition and processing were performed using Analyst TF 1.5 and

Peak View 1.1.1.2 software.

1H), 7.81 (s, 1H), 6.96 (s, 1H), 6.90 (s, 1H), 6.30 (d, J=2.5 Hz, 1H),

4.38 (pent, J=8.3 Hz, 1H), 4.25 (dd, J=8.0, 3.5 Hz, 1H), 3.34 (s, 3H),

2.15–2.05 (m, 1H), 2.05–1.92 (m, 2H), 1.92–1.80 (m, 4H), 1.77–1.60

(m, 3H), 0.86 ppm (t, J=7.3 Hz, 3H); LC–MS m/z: 326.0 [M+H]⁺;

t_R=1.54 min (Method E); HRMS (ESI+) m/z calcd for C₁₈H₂₄N₅O⁺

326.1975, found 326.1978.

(R)-8-Cyclopentyl-7-ethyl-5-methyl-2-(1H-pyrazol-4-yl)-7,8-dihy-

1

dropteridin-6(5H)-one (5): H NMR (500 MHz, CDCl₃): d=8.26 (brs,

2H), 7.86 (s, 1H), 4.37 (t, J=8.3 Hz, 1H), 4.30–4.25 (m, 1H), 3.35 (s,

3H), 2.15–2.05 (m, 1H), 2.05–1.95 (m, 2H), 1.95–1.80 (m, 4H), 1.80–

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 20

&15

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

1.65 (m, 3H), 0.87 ppm (t, J=7.3 Hz, 3H); LC–MS m/z: 327.0 [M+

H]⁺; t_R=1.42 min (Method E); HRMS (ESI+) m/z calcd for

C₁₇H₂₃N₆O⁺327.1928, found 327.1929.

3H), 0.80 ppm (t, J=7.6 Hz, 3H); LC–MS m/z: 377.2 [M+H]⁺; t_R=

4.065 min (Method C); HRMS (ESI+) m/z calcd for C₂₁H₂₅N₆O⁺

377.2084, found 377.2088.

(R)-8-Cyclopentyl-7-ethyl-5-methyl-2-(thiazol-5-yl)-7,8-dihydrop-

teridin-6(5H)-one (6): H NMR (500 MHz, CD₃OD): d=9.03 (s, 1H),

(R)-8-Cyclobutyl-7-ethyl-5-methyl-2-(2-phenyl-1H-imidazol-1-yl)-

7,8-dihydropteridin-6(5H)-one (13): H NMR (300 MHz, CDCl₃): d=

1

8.50 (s, 1H), 7.98 (s, 1H), 4.38 (dd, J=7.5, 3.5 Hz, 1H), 4.30 (pent,

J=8.5 Hz, 1H), 3.40 (s, 3H), 2.25–2.15 (m, 1H), 2.15–1.90 (m, 6H),

1.90–1.80 (m, 1H), 1.80–1.65 (m, 2H), 0.85 ppm (t, J=7.5 Hz, 3H);

LC–MS m/z: 344.1 [M+H]⁺; t_R=2.108 min (Method F); HRMS

(ESI+) m/z calcd for C₁₇H₂₂N₅OS⁺344.1540, found 344.1543.

7.78–7.76 (m, 2H), 7.62–7.60 (m, 1H), 7.57–7.50 (m, 3H), 7.49–7.43

(m, 2H), 4.27–4.24 (m, 1H), 3.69–3.57 (m, 1H), 3.37 (s, 3H), 1.98–

1.40 (m, 8H), 0.78 ppm (t, J=7.5 Hz, 3H); LC–MS m/z: 389.2 [M+

H]⁺; t_R=6.59 min (Method C); HRMS (ESI+) m/z calcd for

C₂₂H₂₅N₆O⁺389.2084, found 389.2088.

(R)-8-Cyclopentyl-7-ethyl-5-methyl-2-(pyridin-4-yl)-7,8-dihydrop-

teridin-6(5H)-one (7): ¹H NMR (500 MHz, CDCl₃): d=8.72 (d, J=

4.5 Hz, 2H), 8.17 (d, J=4.5 Hz, 2H), 7.99 (s, 1H), 4.46 (pent, J=

8.5 Hz, 1H), 4.32 (dd, J=7.8, 3.8 Hz, 1H), 3.40 (s, 3H), 2.23–2.10 (m,

1H), 2.10–1.95 (m, 2H), 1.95–1.80 (m, 4H), 1.80–1.65 (m, 3H),

0.88 ppm (t, J=7.5 Hz, 3H); LC–MS (0.05% TFA) m/z: 338.0 [M+

H]⁺; t_R=1.47 min (Method E); HRMS (ESI+) m/z calcd for

C₁₉H₂₄N₅O⁺338.1975, found 338.1978.

(R)-7-Ethyl-5-methyl-2-(2-phenyl-1H-imidazol-1-yl)-8-((S)-tetrahy-

drofuran-3-yl)-7,8-dihydropteridin-6(5H)-one

(14):

¹H NMR

(400 MHz, CDCl₃): d=7.84 (m, 2H), 7.45–7.62 (m, 6H), 4.30 (dd, J=

7.1, 2.6 Hz, 1H), 3.88–3.94 (m, 2H), 3.55–3.59 (m, 2H), 3.42–3.48 (m,

1H), 3.38 (s, 3H), 1.61–1.91 (m, 4H), 0.82 ppm (t, J=7.5 Hz, 3H);

LC–MS m/z: 405.2 [M+H]⁺; t_R=4.176 min (Method C); HRMS

+

(ESI+) m/z calcd for C₂₂H₂₅N₆O₂405.2034, found 405.2044; Anal.

calcd for C₂₂H₂₄N₆O₂·1.5CF₃COOH: C 52.18, H 4.47, N 14.60, found C

51.91, H 4.38, N 14.71.

(R)-8-Cyclopentyl-7-ethyl-5-methyl-2-(2-methyl-1H-imidazol-1-

yl)-7,8-dihydropteridin-6(5H)-one (8): ¹H NMR (400 MHz, CDCl₃):

d=7.89 (d, J=2.0 Hz, 1H), 7.83 (s, 1H), 7.33 (d, J=2.0 Hz, 1H), 4.42

(t, J=8.0 Hz, 1H), 4.36 (q, J=3.7 Hz, 1H), 3.41 (s, 3H), 3.10 (s, 3H),

2.21–2.08 (brs, 1H), 2.08–1.62 (m, 9H), 0.90 ppm (t, J=7.5 Hz, 3H);

¹³C NMR (100 MHz, CDCl₃): d=163.5, 152.2, 148.5, 145.6, 136.4,

121.9, 119.3, 118.8, 60.1, 59.4, 29.4, 29.0, 28.4, 27.6, 23.5, 23.2, 14.9,

8.9 ppm; LC–MS m/z: 341.2 [M+H]⁺; t_R=6.16 min (Method A);

HRMS (ESI+) m/z calcd for C₁₈H₂₅N₆O⁺341.2084, found 341.2093.

(R)-7-Ethyl-5-methyl-2-(2-phenyl-1H-imidazol-1-yl)-8-((R)-tetrahy-

drofuran-3-yl)-7,8-dihydropteridin-6(5H)-one

(15):

¹H NMR

(400 MHz, CDCl₃): d=7.84 (m, 2H), 7.46–7.62 (m, 6H), 4.19 (dd, J=

7.3, 2.8 Hz, 1H), 3.92–4.05 (m, 2H), 3.59 (dd, J=10.0, 7.8 Hz, 1H),

3.41–3.43 (m, 1H), 3.40 (s, 3H), 3.29 (dd, J=10.0, 6.1 Hz, 1H), 1.62–

1.92 (m, 4H), 0.82 ppm (t, J=7.5 Hz, 3H); LC–MS m/z: 405.2 [M+

H]⁺; t_R=4.396 min (Method C); HRMS (ESI+) m/z calcd for

+

C₂₂H₂₅N₆O₂405.2034, found 405.2030.

(R)-8-Cyclopentyl-7-ethyl-5-methyl-2-(2-phenyl-1H-imidazol-1-

yl)-7,8-dihydropteridin-6(5H)-one (9): ¹H NMR (400 MHz, CDCl₃):

d=7.79 (s, 1H), 7.75 (d, J=1.2 Hz, 1H), 7.65–7.40 (m, 6H), 4.22 (q,

J=3.6 Hz, 1H), 3.61 (t, J=8.1 Hz, 1H), 3.38 (s, 3H), 1.95–1.80 (m,

1H), 1.70–1.50 (m, 5H), 1.50–1.20 (m, 4H), 0.81 ppm (t, J=7.5 Hz,

3H); LC–MS m/z: 403.2 [M+H]⁺; t_R=7.33 min (Method C); HRMS

(ESI+) m/z calcd for C₂₃H₂₇N₆O⁺403.2241, found 403.2247.

(R)-7-Ethyl-2-(2-(3-fluorophenyl)-1H-imidazol-1-yl)-5-methyl-8-

((R)-tetrahydrofuran-3-yl)-7,8-dihydropteridin-6(5H)-one

(16):

¹H NMR (400 MHz, CDCl₃): d=7.89 (d, J=2.0 Hz, 1H), 7.83 (s, 1H),

7.61 (s, 1H), 7.44–7.51 (m, 2H), 7.17–7.25 (m, 2H), 4.01–4.23 (m,

2H), 3.63–3.69 (m, 2H), 3.51 (dd, J=10.1, 3.6 Hz, 1H), 3.38–3.34 (m,

1H), 3.35 (s, 3H), 1.65–1.99 (m, 4H), 0.82 ppm (t, J=7.5 Hz, 3H);

LC–MS m/z: 423.1 [M+H]⁺; t_R=4.522 min (Method C); HRMS

+

(ESI+) m/z calcd for C₂₂H₂₄FN₆O₂423.1939, found 423.1937.

(R)-8-Cyclopentyl-7-ethyl-5-methyl-2-(4-phenyl-1H-imidazol-1-

yl)-7,8-dihydropteridin-6(5H)-one (10): ¹H NMR (400 MHz, CDCl₃):

d=9.34 (s, 1H), 8.15 (s, 1H), 7.79–7.83 (m, 3H), 7.42–7.53 (m, 3H),

4.40–4.48 (m, 1H), 4.34–4.36 (m, 1H), 3.41 (s, 3H), 1.79–2.20 (m,

10H), 0.95 ppm (t, J=7.3 Hz, 3H); LC–MS m/z: 403.1 [M+H]⁺; t_R=

5.049 min (Method A).

(R)-7-Ethyl-2-(2-(4-fluorophenyl)-1H-imidazol-1-yl)-5-methyl-8-

((R)-tetrahydrofuran-3-yl)-7,8-dihydropteridin-6(5H)-one

(17):

¹H NMR (400 MHz, CDCl₃): d=7.90 (s, 1H), 7.81 (s, 1H), 7.57–7.60

(m, 3H), 7.19 (m, 2H), 4.17–4.24 (m, 2H), 3.91–3.99 (m, 1H), 3.66

(dd, J=10.0, 7.9 Hz, 1H), 3.55 (dd, J=9.9, 2.9 Hz, 1H), 3.44–3.46 (m,

1H), 3.39 (s, 3H), 1.66–2.03 (m, 4H), 0.83 ppm (t, J=7.5 Hz, 3H);

LC–MS m/z: 423.2 [M+H]⁺; t_R=4.837 min (Method C); HRMS

(R)-8-Cyclopentyl-7-ethyl-5-methyl-2-(2-(thiazol-4-yl)-1H-imida-

zol-1-yl)-7,8-dihydropteridin-6(5H)-one (11): ¹H NMR (400 MHz,

CDCl₃): d=8.72 (s, 1H), 8.58 (s, 1H), 7.81 (s, 1H), 7.68 (s, 1H), 7.48

(s, 1H), 4.28 (dd, J=7.4, 3.4 Hz, 1H), 3.85 (m, 1H), 3.42 (s, 3H),

1.47–1.94 (m, 10H), 0.90 ppm (t, J=7.4 Hz, 3H); LC–MS m/z: 410.1

[M+H]⁺; t_R=5.353 min (Method C).

+

(ESI+) m/z calcd for C₂₂H₂₄FN₆O₂423.1939, found 423.1942.

(S)-5,6a-Dimethyl-2-(2-phenyl-1H-imidazol-1-yl)-6a,7,8,9-

tetrahydropyrrolo[2,1-h]pteridin-6(5H)-one

(18):

¹H NMR

(400 MHz, CDCl₃): d=7.86 (d, J=2.7 Hz, 1H), 7.73 (s, 1H), 7.62 (d,

J=2.7 Hz, 1H), 7.48 (m, 6H), 3.36 (s, 3H), 3.21 (m, 2H), 2.19 (m,

2H), 2.02 (m, 2H), 1.29 ppm (s, 3H); LC–MS m/: 361.1z [M+H]⁺;

t_R=4.55 min (Method C); HRMS (ESI+) m/z calcd for C₂₀H₂₁N₆O⁺

361.1771, found 361.1768.

(R)-7-Ethyl-8-isopropyl-5-methyl-2-(2-phenyl-1H-imidazol-1-yl)-

7,8-dihydropteridin-6(5H)-one (12): A mixture of intermediate 34

(100 mg, 0.37 mmol), 2-phenyl-1H-imidazole (533 mg, 3.7 mmol),

Cu^I(35 mg, 0.18 mmol), trans-1,2-bis(methylamino)cyclocyclohex-

ane (52.5 mg, 0.07 mL, 0.37 mmol) and solid K₂CO₃(511 mg,

3.7 mmol) in 2 mL DMF was heated in a microwave reaction appa-

ratus for 2 h at 2008C. After this time the reaction was transferred

to a round-bottom flask with the aid of EtOAc, then evaporated.

The residue was purified by reversed-phase (RP) HPLC (PLRPS C₁₈

column, eluting with a gradient of 20–25% CH₃CN in H₂O over

(S)-6a-Ethyl-5-methyl-2-(2-phenyl-1H-imidazol-1-yl)-6a,7,8,9-

tetrahydropyrrolo[2,1-h]pteridin-6(5H)-one (19): The target com-

pound was prepared from 39 and 2-(phenyl)imidazole. The prod-

uct was purified by HPLC (20–40% CH₃CN/H₂O in 30 min,

20 mLmin^ꢀ1, 210 nm, 0.1% TFA, Phenomenex Luna C₁₈, 2ꢂ25 cm);

12 mg of the titled compound was obtained with 99% purity by

LC–MS m/z: 375.2 [M+H]⁺; t_R=5.24 (Method C); ¹H NMR

(400 MHz, CDCl₃): d=7.83 (s, 1H), 7.69 (s, 1H), 7.60–7.42 (m, 6H),

3.25 (s, 3H), 3.21 (m, 1H), 3.14 (m, 1H), 2.00 (m, 2H), 1.90 (m, 2H),

1

30 min) to give the title compound. H NMR (400 MHz, CDCl₃): d=

7.81 (d, J=2.0 Hz, 1H), 7.78 (s, 1H), 7.61 (d, J=2.0 Hz, 1H), 7.50 (m,

5H), 4.29 (dd, J=7.2, 3.2 Hz, 1H), 3.74 (m, 1H), 3.37 (s, 3H), 1.92

(m, 1H), 1.65 (m, 1H), 1.05 (d, J=6.8 Hz, 3H), 0.88 (d, J=6.8 Hz,

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 20

&16

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

+

1.76 (m, 1H), 1.57 (m, 1H), 0.77 ppm (t, J=7.5 Hz, 3H); HRMS

(ESI+) m/z calcd for C₂₁H₂₃N₆O⁺375.1928, found 375.1919.

5.118 min (Method A); HRMS (ESI+) m/z calcd for C₂₃H₂₄N₅O₂

402.1925, found 402.1921.

(S)-6a-Ethyl-5-methyl-2-(2-phenyl-1H-imidazol-1-yl)-7,8,9,10-tet-

rahydro-5H-pyrido[2,1-h]pteridin-6(6aH)-one (20): The racemic

mixture was resolved by chiral HPLC using an isocratic mixture of

EtOH/hexane (1:9, 1 mLmin^ꢀ1) as eluent from a ChiralPak IA

column (4.6ꢂ250 mm, 5 mm). Under these conditions, compound

20 has a retention time of 13.88 min; ¹H NMR (400 MHz, CDCl₃):

d=7.69 (s, 1H), 7.65 (s, 1H), 7.47–7.42 (m, 2H), 7.35–7.28 (m, 3H),

7.17 (s, 1H), 3.59 (d, J=13.4 Hz, 1H), 3.32 (s, 3H), 2.28 (t, J=

11.6 Hz, 1H), 2.21 (d, J=12.6 Hz, 1H), 2.09 (sext, J=7.3 Hz, 1H),

1.86–1.52 (m, 4H), 1.42 (d, J=13.2 Hz, 1H), 1.17–1.03 (m, 1H),

0.67 ppm (t, J=7.4 Hz, 3H); LC–MS m/z: 389.2 [M+H]⁺; t_R=6.85

(Method C); HRMS (ESI+) m/z calcd for C₂₂H₂₅N₆O⁺389.2084,

found 389.2085.

(S)-6a-Ethyl-5-methyl-2-(5-phenyl-2H-1,2,3-triazol-4-yl)-6a,7,9,10-

tetrahydro[1,4]oxazino[3,4-h]pteridin-6(5H)-one (25): To a solution

of 42 (0.247 mmol, 0.07 g) in CH₃CN (2 mL) was added Pd(PPh₃)₄

(0.007 mmol, 0.008 g), phenylacetylene (0.296 mmol, 0.032 mL), Cu^I

(0.007 mmol, 0.001 g), and Et₃N (0.741 mmol, 0.09 mL). The reaction

mixture was microwaved for 25 min at 1408C. The reaction mixture

was filtered and concentrated. The resulting residue was purified

by flash chromatography (30% EtOAc in hexanes). The resulting

residue was dissolved in DMSO (1 mL) and NaN₃(0.071 mmol,

0.005 g) was added. The reaction mixture was microwaved for

30 min at 1758C. The reaction mixture was diluted with EtOAc

(10 mL), washed with H₂O (10 mL), dried with Na₂SO₄, filtered, and

concentrated. The resulting residue was purified by RP-HPLC to

provide 25 as a white solid (1.9 mg, 11%); ¹H NMR (400 MHz,

CDCl₃): d=7.96 (s, 1H), 7.83 (m, 2H), 7.40 (m, 3H), 4.17 (m, 1H),

3.85 (m, 2H), 3.64 (m, 1H), 3.44 (m, 1H), 3.37 (s, 3H), 2.97 (m, 1H),

2.24 (m, 1H), 1.96 (m, 1H), 0.77 ppm (t, J=7.6 Hz, 3H); LC–MS m/z:

392.1 [M+H]⁺; t_R=2.331 min (Method A); HRMS (ESI+) m/z calcd

(S)-6a-Ethyl-5-methyl-2-(2-phenyl-1H-imidazol-1-yl)-6a,7,9,10-

tetrahydro[1,4]oxazino[3,4-h]pteridin-6(5H)-one (21): The purity

of 21 was 99% by LC–MS m/z: 390.1 [M+H]⁺; t_R=5.24 min (Meth-

1

od C); H NMR (400 MHz, CDCl₃): d=7.76 (s, 1H), 7.73 (s, 1H), 7.44

(m, 1H), 7.33 (m, 4H), 7.18 (s, 1H), 4.13 (d, J=11.6 Hz, 1H), 3.72 (d,

J=11.2 Hz, 1H), 3.61 (d, J=11.8 Hz, 1H), 3.35 (s, 3H), 3.23 (m, 2H),

2.63 (m, 1H), 2.23 (m, 1H), 1.91 (m, 1H), 0.74 ppm (t, J=7.4 Hz,

+

for C₂₀H₂₂N₇O₂392.1830, found 392.1823.

(S)-6a-Ethyl-5-methyl-2-(3-phenylisoxazol-4-yl)-6a,7,9,10-

+

3H); HRMS (ESI+) m/z calcd for C₂₁H₂₃N₆O₂391.1877, found

tetrahydro[1,4]oxazino[3,4-h]pteridin-6(5H)-one (26): To a solution

of intermediate 44 (0.2707 mmol, 0.099 g) in 3 mL DMF wad added

N,N-dimethylformamide dimethylacetal (0.8122 mmol, 0.11 mL).

The reaction mixture was stirred at room temperature for 18 h. Hy-

droxylamine hydrochloride (2.707 mmol, 0.629 g) was added, and

the reaction mixture was plunged into a preheated (608C) oil bath

and was stirred for 1 h. The reaction mixture was cooled to room

temperature, diluted with EtOAc, and washed with H₂O. The organ-

ic layer was dried over Na₂SO₄, filtered, and concentrated to give

a racemic mixture of the compounds. The racemate was resolved

by chiral HPLC using an isocratic mixture of EtOH/hexane (3:7,

391.1872.

(R)-6a-Ethyl-5-methyl-2-(2-phenyl-1H-imidazol-1-yl)-6a,7,9,10-

tetrahydro[1,4]oxazino[3,4-h]pteridin-6(5H)-one (ent-21): The

purity of ent-21 was 99% by LC–MS m/z: 390.1 [M+H]⁺; t_R=

5.21 min (Method C); ¹H NMR (400 MHz, CDCl₃): d=7.75 (s, 1H),

7.72 (d, J=2 Hz, 1H), 7.44 (m, 2H), 7.32 (m, 3H), 7.18 (d, J=2H),

4.13 (d, J=15.5 Hz, 1H), 3.71 (dd, J=15.5, 5.2 Hz, 1H), 3.60 (d, J=

15.5 Hz, 1H), 3.34 (s, 3H), 3.25–3.14 (m, 2H), 2.62 (m, 1H), 2.25 (m,

1H), 1.92 (m, 1H), 0.72 ppm (t, J=7.4 Hz, 3H). The absolute config-

uration of ent-21 was assigned based on its Plk-2 activity relative

to the opposite enantiomer 21.

1 mLmin^ꢀ1

) as eluent from a Chiracel OD-H column (0.46ꢂ

1

250 mm) to provide compound 26: H NMR (400 MHz, CDCl₃): d=

8.86 (s, 1H), 7.98 (m, 2H), 7.89 (s, 1H), 7.44 (m, 3H), 4.22 (d, J=

11.6 Hz, 1H), 4.00 (dd, J=13.9, 2.6 Hz, 1H), 3.91 (dd, J=11.6,

3.8 Hz, 1H), 3.66 (d, J=11.6 Hz, 1H), 3.51 (m, 1H), 3.35 (s, 3H), 3.04

(m, 2H), 2.26 (m, 1H), 1.99 (m, 1H), 0.79 ppm (t, J=7.5 Hz, 3H);

LC–MS m/z: 392.1 [M+H]⁺; t_R=4.284 min (Method A); HRMS

(S)-6a-Ethyl-5,8-dimethyl-2-(2-phenyl-1H-imidazol-1-yl)-7,8,9,10-

tetrahydro-5H-pyrazino[2,1-h]pteridin-6(6aH)-one (22): ¹H NMR

(400 MHz, CD₃OD): d=7.89 (s, 1H), 7.78 (s, 1H), 7.45–7.30 (m, 5H),

7.13 (s, 1H), 3.50–3.30 (m, 1H), 3.35 (s, 3H), 3.15 (d, J=11.4 Hz,

1H), 2.62–2.50 (m, 2H), 2.35–2.25 (m, 1H), 2.25 (s, 3H), 2.13 (d, J=

11.6 Hz, 1H), 1.90–1.78 (m, 1H), 1.72 (t, J=11.9 Hz, 1H), 0.67 ppm

(t, J=7.4 Hz, 3H); LC–MS m/z: 404.2 [M+H]⁺; t_R=13.58 min (Chir-

alPak IA 10); (ꢀ) rotating enantiomer; HRMS (ESI+) m/z calcd for

C₂₂H₂₆N₇O⁺404.2193, found 404.2186.

+

(ESI+) m/z calcd for C₂₁H₂₂N₅O₃392.1718, found 392.1711.

(S)-6a-Ethyl-5-methyl-2-(4-phenyl-1,2,3-thiadiazol-5-yl)-6a,7,9,10-

tetrahydro[1,4]oxazino[3,4-h]pteridin-6(5H)-one (27): To a solution

of intermediate 44 (0.2109 mmol, 0.074 g) in 2 mL EtOH, hydrazine

(0.707 mmol, 0.023 mL) was added. The reaction mixture was

plunged into a preheated (808C) oil bath and was stirred for 18 h.

The reaction mixture was cooled to room temperature and con-

centrated. Thionyl chloride (2 mL) was slowly added to the residue.

The reaction mixture was stirred for 15 min and then concentrated.

The crude mass was dissolved in 10 mL CH₂Cl₂and washed with

saturated aqueous NaHCO₃, dried over Na₂SO₄, filtered, and con-

centrated to give a racemic mixture of the compounds. The race-

mate was resolved by chiral HPLC using an isocratic mixture of

EtOH/hexane (20:80; 1 mLmin^ꢀ1) as eluent with a Chiracel IA 4.6ꢂ

250 mm column to give isolated 27 and ent-27. Characterization

for 27: ¹H NMR (400 MHz, CDCl₃): d=7.83 (s, 1H), 7.74 (m, 2H), 7.44

(m, 3H), 4.14 (d, J=11.2 1.2 Hz, 1H), 3.84 (m, 1H), 3.62 (m, 2H),

3.40 (m, 1H), 3.35 (s, 3H), 2.84 (m, 1H), 2.23 (m, 1H), 1.97 (m, 1H),

0.76 ppm (t, J=7.4 Hz, 3H) LC–MS m/z: 409.0 [M+H]⁺; t_R=

6.91 min (Method A); HRMS (ESI+) m/z calcd for C₂₀H₂₁N₆O₂S⁺

409.1447, found 409.1433.

(S)-6a-Ethyl-5-methyl-2-(3-phenylpyridin-4-yl)-6a,7,9,10-

tetrahydro[1,4]oxazino[3,4-h]pteridin-6(5H)-one (23): ¹H NMR

(400 MHz, CDCl₃): d=8.69 (d, J=6.8 Hz, 1H), 8.67 (s, 1H), 7.89 (m,

2H), 7.30 (m, 5H), 7.20 (m, 2H), 4.12 (d, J=15.4 Hz, 1H), 3.71 (dd,

J=14.9, 4.9 Hz, 1H), 3.54 (d, J=15.4 Hz, 1H), 3.35 (s, 3H), 3.20 (m,

2H), 2.55 (m, 1H), 2.15 (m, 1H), 1.85 (m, 1H), 0.67 ppm (t, J=

10.1 Hz, 3H); LC–MS m/z: 402.1 [M+H]⁺; t_R=5.886 min (Meth-

+

od A); HRMS (ESI+) m/z calcd for C₂₃H₂₃N₅O₂402.1925, found

402.1920.

(S)-6a-Ethyl-5-methyl-2-(2-phenylpyridin-3-yl)-6a,7,9,10-

tetrahydro[1,4]oxazino[3,4-h]pteridin-6(5H)-one (24): ¹H NMR

(400 MHz, CDCl₃): d=8.74 (dd, J=4.8, 1.7 Hz, 1H), 8.28 (dd, 7.8,

1.7 Hz, 1H), 7.92 (s, 1H), 7.32 (m, 3H), 7.28 (m, 10H), 4.09 (d, J=

11.6 Hz, 1H), 3.72 (dd, J=11.5, 3.7 Hz, 1H), 3.56 (d, J=11.6 Hz, 1H),

3.35 (s, 3H), 3.22 (m, 2H), 2.59 (m, 1H), 2.16 (m, 1H), 1.87 (m, 1H),

0.69 ppm (t, J=7.5 Hz, 3H); LC–MS m/z: 402.1 [M+H]⁺; t_R=

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 20

&17

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

(S)-2-(2-(2,4-Difluorophenyl)-1H-imidazol-1-yl)-6a-ethyl-5-methyl-

6a,7,9,10-tetrahydro[1,4]oxazino[3,4-h]pteridin-6(5H)-one (28):

¹H NMR (400 MHz, CDCl₃): d=7.83 (d, J=1.5 Hz, 1H), 7.72 (s, 1H),

7.67 (m, 1H), 7.22 (d, J=1.5 Hz, 1H), 6.99 (m, 1H), 6.75 (m, 1H),

4.16 (d. J=11.7 Hz, 1H), 3.80 (m, 1H), 3.61 (d, J=11.7 Hz, 1H), 3.34

(m, 5H), 2.76 (m, 1H), 2.26 (m, 1H), 1.94 (m, 1H), 0.75 ppm (t, J=

7.5 Hz, 3H); LC–MS m/z: 427.1 [M+H]⁺; t_R=2.334 min (Method A);

compound (6.16 g, 61%); LC–MS m/z: 315.0 [M+H]⁺; t_R=

6.093 min (Method C); ¹H NMR (400 MHz, CDCl₃): d=8.73 (s, 1H),

3.73 (s, 3H), 3.61 (m, 1H), 2.99 (m, 1H), 2.61 (m, 1H), 2.24–2.04 (m,

5H), 0.89 ppm (t, J=7.4 Hz, 3H).

(S)-2-Chloro-6a-ethyl-5-methyl-6a,7,8,9-tetrahydropyrrolo[2,1-

h]pteridin-6(5H)-one (39): ¹H NMR (300 MHz, CDCl₃): d=7.63

(s,1H), 3.85 (m, 1H), 3.71 (m, 1H), 3.33 (s, 3H), 2.28 (m, 2H), 2.04

(m, 2H), 1.78 (m,1H), 1.62 (m, 1H), 0.80 ppm (t, J=7.5 Hz, 3H); LC–

MS m/z: 267.0 [M+H]⁺; t_R=2.781 min (Method A).

+

HRMS (ESI+) m/z calcd for C₂₁H₂₁F₂N₆O₂427.1689, found

427.1680.

(S)-2-(2-(2,3-Difluorophenyl)-1H-imidazol-1-yl)-6a-ethyl-5-methyl-

6a,7,9,10-tetrahydro[1,4]oxazino[3,4-h]pteridin-6(5H)-one (29):

¹H NMR (400 MHz, CD₃OD): d=7.96 (d, J=1.6 Hz, 1H), 7.91 (s, 1H),

7.44–7.34 (m, 2H), 7.33–7.25 (m, 1H), 7.21 (d, J=1.6 Hz, 1H), 4.02

(d, J=11.6 Hz, 1H), 3.75 (dd, J=11.4, 4.0 Hz, 1H), 3.64 (d, J=

11.6 Hz, 1H), 3.37 (dd, J=11.8, 3.2 Hz, 1H), 3.34 (s, 3H), 2.80–2.60

(m, 1H), 2.31 (sext, J=7.5 Hz, 1H), 1.86 (sext, J=7.5 Hz, 1H),

0.72 ppm (t, J=7.6 Hz, 3H); LC–MS m/z: 427.1 [M+H]⁺; t_R=

6.31 min (Method C); t_R=10.50 min (ChiralPak IA 10); (+) rotating

Methyl 4-(2-chloro-5-nitropyrimidin-4-yl)-3-ethylmorpholine-3-

carboxylate (41): ¹H NMR (400 MHz, CDCl₃): d=8.78 (s, 1H), 3.87

(m, 4H), 3.75 (s, 3H), 3.54 (m, 1H), 3.04 (m, 1H), 2.48 (m, 1H), 1.97

(m, 1H), 0.86 ppm (t, J=7.3 Hz, 3H); LC–MS m/z: 331.0 [M+H]⁺;

t_R=4.918 min (Method A).

2-Chloro-6a-ethyl-5-methyl-6a,7,9,10-tetrahydro[1,4]oxazino[3,4-

h]pteridin-6(5H)-one (42): ¹H NMR (400 MHz, CDCl₃): d=7.76 (s,

1H), 4.34 (dd, J=13.9, 2.8 Hz, 1H), 4.18 (d, J=11.7 Hz, 1H), 4.01

(dd, J=11.7, 3.9 Hz, 1H), 3.66 (d, J=11.7 Hz, 1H), 3.57 (dt, J=12.2,

3.1 Hz, 1H), 3.32 (s, 3H), 3.23 (m, 1H), 2.32 (m, 1H), 2.01 (m, 1H),

0.79 ppm (t, J=7.6 Hz, 3H); LC–MS m/z: 282.9 [M+H]⁺; t_R=

2.717 min (Method A).

+

enantiomer; HRMS (ESI+) m/z calcd for C₂₁H₂₁F₂N₆O₂427.1689,

found 427.1682.

(S)-6a-Ethyl-2-(2-(5-fluoropyridin-2-yl)-1H-imidazol-1-yl)-5-

methyl-6a,7,9,10-tetrahydro[1,4]oxazino[3,4-h]pteridin-6(5H)-

one (30): ¹H NMR (400 MHz, CDCl₃): d=8.54 (dd, J=6.4, 1.3 Hz,

1H), 8.39 (d, J=2.3 Hz, 1H), 7.87 (d, J=1.9 Hz, 1H), 7.72 (s, 1H),

7.67 (m, 1H), 7.28 (d, J=1.9 Hz, 1H), 4.12 (d, J=15.5 Hz, 1H), 3.78

(dd, J=15.3, 5.3 Hz, 1H), 3.63 (d, 15.6 Hz, 1H), 3.31 (m, 5H), 2.71

(m, 1H), 2.25 (m, 1H), 1.93 (m, 1H), 0.75 ppm (t, J=10.0 Hz, 3H);

LC–MS m/z: 410.1 [M+H]⁺; t_R=4.775 min (Method A); HRMS

6a-Ethyl-5-methyl-2-(2-oxo-2-phenylethyl)-6a,7,9,10-tetrahydro-

[1,4]oxazino[3,4-h]pteridin-6(5H)-one (44): Intermediate 42

(0.707 mmol, 0.200 g), sodium methanethiolate (2.12 mmol,

0.148 g) and THF (2 mL) were combined in a sealed tube and

heated at 1208C for 18 h. The reaction mixture was cooled to

room temperature, diluted with 15 mL EtOAc, washed with H₂O,

dried with Na₂SO₄, filtered and concentrated. The resulting residue

was dissolved in 2 mL AcOH, the temperature was decreased to

08C, and a solution of KMnO₄(0.848 mmol, 0.134 g) in 2 mL H₂O

was added. The reaction mixture was stirred for 2 h at 08C, then

was quenched with saturated sodium thiosulfate and warmed to

room temperature and extracted into EtOAc (3ꢂ15 mL). The com-

bined organic layers were dried with Na₂SO₄, filtered, and concen-

trated. The resulting residue was purified by flash chromatography

(50% EtOAc in hexanes) to give 6a-ethyl-5-methyl-2-(methylsulfon-

yl)-6a,7,9,10-tetrahydro[1,4]oxazino[3,4-h]pteridin-6(5H)-one.

+

(ESI+) m/z calcd for C₂₀H₂₁FN₇O₂410.1735, found 410.1727.

(S)-6a-Ethyl-5-methyl-2-(2-(thiazol-2-yl)-1H-imidazol-1-yl)-

6a,7,9,10-tetrahydro[1,4]oxazino[3,4-h]pteridin-6(5H)-one (31):

¹H NMR (400 MHz, CDCl₃): d=7.80 (s, 1H), 7.73 (s, 1H), 7.65 (s, 1H),

7.39 (m, 1H), 7.22 (m, 1H), 4.16 (d, J=11.6 Hz, 1H), 3.85 (m, 1H),

3.69 (m, 1H), 3.64 (m, 1H), 3.44 (m, 1H), 3.37 (s, 3H), 2.90 (m, 1H),

2.28 (m, 1H), 1.97 (m, 1H), 0.80 ppm (t, J=7.5 Hz, 3H); LC–MS m/z:

398.1 [M+H]⁺; t_R=4.766 min (Method A); HRMS (ESI+) m/z calcd

for C₁₈H₂₀N₇O₂S⁺398.1394, found 398.1385.

(R)-2-Chloro-8-cyclobutyl-7-ethyl-5-methyl-7,8-dihydropteridin-

6(5H)-one (33): ¹H NMR (CDCl₃, 300 MHz): d=7.67 (s, 1H), 4.42–

4.36 (m, 1H), 4.30–4.27 (m, 1H), 3.32 (s, 3H), 2.58–2.56 (m, 1H),

2.27–2.02 (m, 3H), 1.81–1.58 (m, 4H), 0.84 ppm (t, J=4.3 Hz, 3H);

LC–MS m/z: 281.0 [M+H]⁺; t_R=9.47 min (Method B).

The above intermediate was added to a suspension of NaH

(1.81 mmol, 0.07 g) and acetophenone (1.64 mmol, 0.191 g) in

3 mL THF with stirring at 08C. The reaction mixture was stirred for

18 h while slowly warming to room temperature. The reaction mix-

ture was quenched with 10 mL saturated NH₄Cl, diluted with

20 mL of EtOAc, and the two layers were separated. The organic

layer was dried with Na₂SO₄, filtered, and concentrated. The result-

ing residue was purified by flash chromatography (50% EtOAc in

hexanes) to provide the title compound as a white solid (0.07 g,

62%): ¹H NMR (400 MHz, CDCl₃): d=7.99 (d, J=7.6 Hz, 1H), 7.81

(m, 2H), 7.40 (m, 3H), 4.39 (s, 1H), 4.19 (m, 1H), 3.95 (m, 1H), 3.61

(m, 1H), 3.50 (m, 1H), 3.32 (s, 3H), 2.19 (m, 1H), 1.93 (m, 1H),

0.79 ppm (t, J=7.6 Hz, 3H); LC–MS m/z: 376.2 [M+H]⁺; t_R=

2.308 min (Method A).

(R)-2-Chloro-7-ethyl-5-methyl-8-((R)-tetrahydrofuran-3-yl)-7,8-di-

hydropteridin-6(5H)-one (35): ¹H NMR (400 MHz, CDCl₃): d=7.72

(s, 1H), 4.79 (m, 1H), 4.23 (dd, J=8.0, 3.8 Hz, 1H), 4.15 (m, 1H),

3.94 (d, J=5.6 Hz, 2H), 3.87 (m, 1H), 3.33 (s, 3H), 2.28 (m, 1H), 2.19

(m, 1H), 1.92 (m, 1H), 1.73 (m, 1H), 0.88 ppm (t, J=7.5 Hz, 3H);

LC–MS m/z: 345.1 [M+H]⁺; t_R=5.312 min (Method A).

(S)-Methyl 1-(2-chloro-5-nitropyrimidin-4-yl)-2-ethylpyrrolidine-

2-carboxylate (38): (S)-Methyl 2-ethylpyrrolidine-2-carboxylate

(6.16 g, 31.93 mmol) was suspended in THF (100 mL) and the tem-

perature was decreased to 08C. N,N-Diisopropylethylamine (DIPEA;

11.08 mL, 67.05 mmol) was added, and the reaction mixture was

stirred for 10 min; 2,4-dichloro-5-nitropyrimidine (6.19 g,

31.93 mmol) was added to the reaction mixture, and the reaction

was stirred at 08C until complete consumption was observed by

TLC. The reaction mixture was diluted with EtOAc and washed

with H₂O. The organic layer was dried with anhydrous Na₂SO₄, fil-

tered, and concentrated. The resulting residue was purified by

flash chromatography (20% EtOAc in hexanes) to provide title

Supporting Information

Detailed procedures for the synthesis of compounds 51, (ꢁ)-22,

(ꢁ)-64, and (ꢁ)-7, detailed procedures for the in vitro assays, as

well as a full listing of the panel of kinases for inhibition assays by

compound 21 are provided in the Supporting Information.

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 20

&18

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

Angelis, J. Seibler, J. Yuan, M. Bergmann, R. Knecht, B. Kreft, K. Streb-

hardt, Nat. Commun. 2011 , 2 , 395.

Acknowledgements

[14] a) D. P. Seeburg, M. Feliu-Mojer, J. Gaiottino, D. T. Pak, M. Sheng, Neuron

2008 , 58 , 57 1 –5 83; b) G. Kauselmann, M. Weiler, P. Wulff, S. Jessberger,

U. Konietzko, J. Scafidi, U. Staubli, J. Bereiter-Hahn, K. Strebhardt, D.

Kuhl, EMBO J. 1999 , 18 , 552 8 –5 539.

[15] L. X. Wu, C. K. Sun, Y. M. Zhang, M. Fan, J. Xu, H. Ma, J. Zhang, Brain Res.

2007 , 1168 , 3 8 –4 5.

We gratefully acknowledge Pam Santiago, Christopher Willits,

David Chian, and Erich Goldbach for their skillful support. We es-

pecially thank John Erve for the determination of HRMS data for

all final compounds.

[16] D. M. Evers, J. A. Matta, H. S. Hoe, D. Zarkowsky, S. H. Lee, J. T. Isaac, D. T.

Pak, Nat. Neurosci. 2010 , 13 , 119 9 –1 207.

[17] W. C. Zimmerman, R. L. Erikson, Cell Cycle 2007 , 6 , 131 4 –1 318.

[18] D. T. Pak, M. Sheng, Science 2003 , 302 , 136 8 –1 373.

Keywords: computational chemistry

medicinal chemistry · Parkinson’s disease · polo-like kinase-2

· kinase inhibitors ·

[19] S. Bowers, A. P. Truong, M. Ye, D. L. Aubele, J. M. Sealy, R. J. Neitz, R. K.

Hom, W. Chan, M. S. Dappen, R. A. Galemmo, Jr., A. W. Konradi, H. L.

Sham, Y. L. Zhu, P. Beroza, G. Tonn, H. Zhang, J. Hoffman, R. Motter, D.

Fauss, P. Tanaka, M. P. Bova, Z. Ren, D. Tam, L. Ruslim, J. Baker, D.

Pandya, L. Diep, K. Fitzgerald, D. R. Artis, J. P. Anderson, M. Bergeron,

Bioorg. Med. Chem. Lett. 2013 , 23 , 274 3 –2 749.

[20] a) “Pyridodihydropyraziones, process for their manufacture and use

thereof as medicaments”: T. Brandl, M. Grauert, R. Hauptmann, M. Hoff-

mann, M. Steegmaier, (Boehringer Ingelheim International GmbH),

US7625899 B2, 2009; b) “Dihydropteridinones, method for the produc-

tion and use thereof in the form of drugs”: A. Baum, T. Brandl, S. Breit-

felder, F. Colbatzky, C. Eickmeier, M. Grauert, M. Hoffmann, J. J. Quant,

G. Schnapp, F. Solca, M. Steegmaier, (Boehringer Ingelheim Pharma

GmbH & Co. Kg), Int. Pat. Appl. WO2004076454 A1, 2004; c) “Method

for the Production of Dihydropteridinones”: G. Linz, G. F. Kraemer, L.

Gutschera, G. Asche, (Boehringer Ingelheim International GmbH), Int.

Pat. Appl. WO 2006018220A2, 2006; d) G. Budin, K. S. Yang, T. Reiner, R.

Weissleder, Angew. Chem. 2011 , 123 , 955 0 –9 55 3; Angew. Chem. Int. Ed.

2011 , 50 , 937 8 –9 381.

[1] E. Kçvari, J. Horvath, C. Bouras, Brain Res. Bull. 2009 , 80 , 20 3 –2 10.

[2] a) H. Fujiwara, M. Hasegawa, N. Dohmae, A. Kawashima, E. Masliah,

M. S. Goldberg, J. Shen, K. Takio, T. Iwatsubo, Nat. Cell Biol. 2002, 4,

160–164; b) Y. Saito, A. Kawashima, N. N. Ruberu, H. Fujiwara, S.

Koyama, M. Sawabe, T. Arai, H. Nagura, H. Yamanouchi, M. Hasegawa, T.

Iwatsubo, S. Murayama, J. Neuropathol. Exp. Neurol. 2003, 62, 644–654;

c) J. P. Anderson, D. E. Walker, J. M. Goldstein, R. de Laat, K. Banducci,

R. J. Caccavello, R. Barbour, J. Huang, K. Kling, M. Lee, L. Diep, P. S. Keim,

X. Shen, T. Chataway, M. G. Schlossmacher, P. Seubert, D. Schenk, S.

Sinha, W. P. Gai, T. J. Chilcote, J. Biol. Chem. 2006 , 281 , 2973 9 –2 9752.

[3] K. J. Inglis, D. Chereau, E. F. Brigham, S. S. Chiou, S. Schobel, N. L. Frigon,

M. Yu, R. J. Caccavello, S. Nelson, R. Motter, S. Wright, D. Chian, P. Santia-

go, F. Soriano, C. Ramos, K. Powell, J. M. Goldstein, M. Babcock, T. Yed-

nock, F. Bard, G. S. Basi, H. Sham, T. J. Chilcote, L. McConlogue, I. Gris-

wold-Prenner, J. P. Anderson, J. Biol. Chem. 2009, 284, 2598–2602.

[4] M. K. Mbefo, K. E. Paleologou, A. Boucharaba, A. Oueslati, H. Schell, M.

Fournier, D. Olschewski, G. Yin, M. Zweckstetter, E. Masliah, P. J. Kahle,

H. Hirling, H. A. Lashuel, J. Biol. Chem. 2010 , 285 , 280 7 –2 822.

[21] M. Kothe, D. Kohls, S. Low, R. Coli, G. R. Rennie, F. Feru, C. Kuhn, Y. H.

Ding, Chem. Biol. Drug Des. 2007 , 70 , 54 0 –5 46.

[22] C. Bissantz, B. Kuhn, M. Stahl, J. Med. Chem. 2010 , 53 , 506 1 –5 084.

[23] W. B. Jennings, B. M. Farrell, J. F. Malone, Acc. Chem. Res. 2001 , 34 , 88 5 –

894.

[5] a) M. Takahashi, H. Kanuka, H. Fujiwara, A. Koyama, M. Hasegawa, M.

Miura, T. Iwatsubo, Neurosci. Lett. 2003 , 336 , 15 5 –1 58; b) M. Neumann,

P. J. Kahle, B. I. Giasson, L. Ozmen, E. Borroni, W. Spooren, V. Muller, S.

Odoy, H. Fujiwara, M. Hasegawa, T. Iwatsubo, J. Q. Trojanowski, H. A.

Kretzschmar, C. Haass, J. Clin. Invest. 2002, 110, 1429–1439; c) P. J.

Kahle, M. Neumann, L. Ozmen, V. Muller, H. Jacobsen, W. Spooren, B.

Fuss, B. Mallon, W. B. Macklin, H. Fujiwara, M. Hasegawa, T. Iwatsubo,

H. A. Kretzschmar, C. Haass, EMBO Rep. 2002 , 3 , 58 3 –5 88.

[6] B. C. van de Weerdt, R. H. Medema, Cell Cycle 2006, 5, 853–864.

[7] a) V. Archambault, D. M. Glover, Nat. Rev. Mol. Cell Biol. 2009 , 10 , 26 5 –

275; b) K. Strebhardt, Nat. Rev. Drug Discovery 2010 , 9 , 64 3 –6 60.

[8] a) E. F. Johnson, K. D. Stewart, K. W. Woods, V. L. Giranda, Y. Luo, Bio-

chemistry 2007 , 46 , 955 1 –9 563; b) M. Steegmaier, M. Hoffmann, A.

Baum, P. Lenart, M. Petronczki, M. Krssak, U. Gurtler, P. Garin-Chesa, S.

Lieb, J. Quant, M. Grauert, G. R. Adolf, N. Kraut, J. M. Peters, W. J. Rettig,

Curr. Biol. 2007 , 17 , 31 6 –3 22; c) K. Strebhardt, A. Ullrich, Nat. Rev.

Cancer 2006 , 6 , 32 1 –3 30.

[9] a) M. Kothe, D. Kohls, S. Low, R. Coli, A. C. Cheng, S. L. Jacques, T. L.

Johnson, C. Lewis, C. Loh, J. Nonomiya, A. L. Sheils, K. A. Verdries, T. A.

Wynn, C. Kuhn, Y. H. Ding, Biochemistry 2007 , 46 , 596 0 –5 971; b) T. M.

Bandeiras, R. C. Hillig, P. M. Matias, U. Eberspaecher, J. Fanghanel, M.

Thomaz, S. Miranda, K. Crusius, V. Putter, P. Amstutz, M. Gulotti-Georgie-

va, H. K. Binz, C. Holz, A. A. Schmitz, C. Lang, P. Donner, U. Egner, M. A.

Carrondo, B. Muller-Tiemann, Acta Crystallogr. Sect. D 2008 , 64 , 33 9 –

353.

[10] Z. Andrysik, W. Z. Bernstein, L. Deng, D. L. Myer, Y. Q. Li, J. A. Tischfield,

P. J. Stambrook, El M. Bahassi, Nucleic Acids Res. 2010 , 38 , 293 1 –2 943.

[11] a) S. Ma, J. Charron, R. L. Erikson, Mol. Cell Biol. 2003 , 23 , 693 6 –6 94 3;

b) Y. Yang, J. Bai, R. Shen, S. A. Brown, E. Komissarova, Y. Huang, N.

Jiang, G. F. Alberts, M. Costa, L. Lu, J. A. Winkles, W. Dai, Cancer Res.

2008 , 68 , 407 7 –4 085.

[24] A. P. Truong, D. L. Aubele, G. D. Probst, M. L. Neitzel, C. M. Semko, S.

Bowers, D. Dressen, R. K. Hom, A. W. Konradi, H. L. Sham, A. W. Garofalo,

P. S. Keim, J. Wu, M. S. Dappen, K. Wong, E. Goldbach, K. P. Quinn, J. M.

Sauer, E. F. Brigham, W. Wallace, L. Nguyen, S. S. Hemphill, M. P. Bova, F.

Bard, T. A. Yednock, G. Basi, Bioorg. Med. Chem. Lett. 2009 , 19 , 492 0 –

4923.

[25] a) J. Srinivasan, T. E. Cheatham, P. Cieplak, P. A. Kollman, D. A. Case, J.

Am. Chem. Soc. 1998 , 120 , 940 1 –9 40 9; b) Y. N. Vorobjev, J. C. Almagro,

J. Hermans, Proteins 1998 , 32 , 39 9 –4 13; c) B. Jayaram, D. Sprous, M. A.

Young, D. L. Beveridge, J. Am. Chem. Soc. 1998 , 120 , 1062 9 –1 0633;

d) P. A. Kollman, I. Massova, C. Reyes, B. Kuhn, S. Huo, L. Chong, M. Lee,

T. Lee, Y. Duan, W. Wang, O. Donini, P. Cieplak, J. Srinivasan, D. A. Case,

T. E. Cheatham III, Acc. Chem. Res. 2000 , 33 , 88 9 –8 97.

[26] While the absolute configuration of the ethyl group at position 7 of the

pteridinone ring has been firmly established by the X-ray co-crystal

structures and precedence of BI 2536, the absolute stereochemical con-

figurations shown for the tetrahydrofuran ring appended to the 8-posi-

tion of 13, and 14–16 had to be assigned arbitrarily. All attempts to de-

termine the absolute stereochemical configuration of the tetrahydrofur-

an ring in 14, and 15–17 or one of the synthetic intermediates by

single-crystal X-ray crystallography were unsuccessful.

[27] a) S. V. Ley, A. W. Thomas, Angew. Chem. 2003 , 115 , 555 8 –5 607; Angew.

Chem. Int. Ed. 2003 , 42 , 540 0 –5 449; b) P. Wang, C. Liang, M. Leung, Tet-

rahedron 2005 , 61 , 293 1 –2 939.

[28] G. D. I. Artman, R. J. Rafferty, R. M. Williams, Org. Synth. 2009, 86, 262–

273.

[12] a) S. Warnke, S. Kemmler, R. S. Hames, H. L. Tsai, U. Hoffmann-Rohrer,

A. M. Fry, I. Hoffmann, Curr. Biol. 2004 , 14 , 120 0 –1 207; b) T. F. Burns, P.

Fei, K. A. Scata, D. T. Dicker, W. S. El-Deiry, Mol. Cell Biol. 2003 , 23 , 555 6 –

5571.

[13] a) T. Matsumoto, P.-y. Wang, W. Ma, H. J. Sung, S. Matoba, P. M. Hwang,

Proc. Natl. Acad. Sci. USA 2009 , 106 , 1454 2 –1 4546; b) M. Raab, S.

Kappel, A. Kramer, M. Sanhaji, Y. Matthess, E. Kurunci-Csacsko, J. Calza-

da-Wack, B. Rathkolb, J. Rozman, T. Adler, D. H. Busch, I. Esposito, H.

Fuchs, V. Gailus-Durner, M. Klingenspor, E. Wolf, N. Sanger, F. Prinz, M. H.

[29] J. J. Jagodzinski, D. L. Aubele, D. A. Quincy, M. S. Dappen, R. K. Hom,

R. A. Galemmo, Jr., A. W. Konradi, H. L. Sham, Tetrahedron Lett. 2011 , 52 ,

247 1 –2 472.

[30] R. Vancraenenbroeck, E. Lobbestael, M. D. Maeyer, V. Baekelandt, J. M.

Taymans, CNS Neurol. Disord. Drug Targets 2011, 10, 724–740.

[31] M. K. Lee in Parkinson’s Disease: Genetics and Pathogenesis (Ed.: T. M.

Dawson), Informa Healthcare USA Inc., 2007, pp. 251–268.

[32] a) D. Kirik, A. Bjorklund, Trends Neurosci. 2003 , 26 , 38 6 –3 92; b) M. F.

Chesselet, Exp. Neurol. 2008 , 209 , 2 2 –2 7.

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 20

&19

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

[33] “Processing of X-ray Diffraction Data Collected in Oscillation Mode”: Z.

Otwinowski, W. Minor in Methods in Enzymology, Vol. 276 (Eds.: C. W.

Carter, Jr., R. M. Sweet), Academic Press, New York, 1997, pp. 307–326.

[34] “Coot News”: B. Lohkamp, P. Emsley, K. Cowtan in CCP4 Newsletter

on Protein Crystallography, No. 42, 2005: http://www.ccp4.ac.uk/

newsletters/newsletter42.pdf (accessed May 26, 2013).

V. Babin, P. A. Kollman, AMBER 10, University of California, San Francisco,

2008.

[37] J. Wang, W. Wang, P. A. Kollman, D. A. Case, J. Mol. Graphics Modell.

2006 , 25 , 24 7 –2 60.

[38] a) M. E. Voss, C. M. Beer, S. A. Mitchell, P. A. Blomgren, P. E. Zhichkin, Tet-

rahedron 2008 , 64 , 64 5 –6 51; b) R. P. Frutos, I. Gallou, D. Reeves, Y. Xu,

D. Krishnamurthy, C. H. Senanayake, Tetrahedron Lett. 2005 , 46 , 836 9 –

8372.

[35] G. N. Murshudov, A. A. Vagin, E. J. Dodson, Acta Crystallogr. Sect. D

1997 , 53 , 24 0 –2 55.

[36] D. A. Case, T. A. Darden, T. E. Cheatham III, C. L. Simmerling, J. Wang,

R. E. Duke, R. Luo, M. Crowley, R. C. Walker, W. Zhang, K. M. Merz, B.

Wang, S. Hayik, A. Roitberg, G. Seabra, I. Kolossvꢃry, K. F. Wong, F. Paesa-

ni, J. Vanicek, X. Wu, S. R. Brozell, T. Steinbrecher, H. Gohlke, L. Yang, C.

Tan, J. Mongan, V. Hornak, G. Cui, D. H. Mathews, M. G. Seetin, C. Sagui,

Received: April 16, 2013

Published online on && &&, 0000

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 20

&20

&

ÞÞ

These are not the final page numbers!

Article Doi

Selective and brain-permeable polo-like kinase-2 (Plk-2) inhibitors that reduce α-synuclein phosphorylation in rat brain

DOI: 10.1002/cmdc.201300166

Source and publish data:

Authors:

Article abstract of DOI:10.1002/cmdc.201300166

Full text of DOI:10.1002/cmdc.201300166

Products guided by the article

R&D Labs maybe for 1313519-76-4

Relevant to this article

Hot Product